Merge commit '88bd7fdc821aaa0cbcf44cf075c62aaa42121e3f'

* commit '88bd7fdc821aaa0cbcf44cf075c62aaa42121e3f':
  Drop DCTELEM typedef

Conflicts:
	libavcodec/alpha/dsputil_alpha.h
	libavcodec/alpha/motion_est_alpha.c
	libavcodec/arm/dsputil_init_armv6.c
	libavcodec/bfin/dsputil_bfin.h
	libavcodec/bfin/pixels_bfin.S
	libavcodec/cavs.c
	libavcodec/cavsdec.c
	libavcodec/dct-test.c
	libavcodec/dnxhdenc.c
	libavcodec/dsputil.c
	libavcodec/dsputil.h
	libavcodec/dsputil_template.c
	libavcodec/eamad.c
	libavcodec/h264_cavlc.c
	libavcodec/h264idct_template.c
	libavcodec/mpeg12.c
	libavcodec/mpegvideo.c
	libavcodec/mpegvideo.h
	libavcodec/mpegvideo_enc.c
	libavcodec/ppc/dsputil_altivec.c
	libavcodec/proresdsp.c

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

21 files changed, 94 insertions, 93 deletions

diff --git a/libavcodec/x86/dnxhdenc.c b/libavcodec/x86/dnxhdenc.c
index b2ba894bf3..98b8b17e16 100644
--- a/libavcodec/x86/dnxhdenc.c
+++ b/libavcodec/x86/dnxhdenc.c
@@ -26,7 +26,7 @@
 
 #if HAVE_SSE2_INLINE
 
-static void get_pixels_8x4_sym_sse2(DCTELEM *block, const uint8_t *pixels, int line_size)
+static void get_pixels_8x4_sym_sse2(int16_t *block, const uint8_t *pixels, int line_size)
 {
     __asm__ volatile(
         "pxor %%xmm5,      %%xmm5       \n\t"
diff --git a/libavcodec/x86/dsputil_mmx.c b/libavcodec/x86/dsputil_mmx.c
index 858eb90773..2bbe7ff9d3 100644
--- a/libavcodec/x86/dsputil_mmx.c
+++ b/libavcodec/x86/dsputil_mmx.c
@@ -226,10 +226,10 @@ DECLARE_ALIGNED(16, const double, ff_pd_2)[2] = { 2.0, 2.0 };
 /***********************************/
 /* standard MMX */
 
-void ff_put_pixels_clamped_mmx(const DCTELEM *block, uint8_t *pixels,
+void ff_put_pixels_clamped_mmx(const int16_t *block, uint8_t *pixels,
                                int line_size)
 {
-    const DCTELEM *p;
+    const int16_t *p;
     uint8_t *pix;
 
     /* read the pixels */
@@ -301,7 +301,7 @@ void ff_put_pixels_clamped_mmx(const DCTELEM *block, uint8_t *pixels,
     "movq               %%mm3, (%0, %3, 2)  \n\t"           \
     "movq               %%mm4, (%0, %1)     \n\t"
 
-void ff_put_signed_pixels_clamped_mmx(const DCTELEM *block, uint8_t *pixels,
+void ff_put_signed_pixels_clamped_mmx(const int16_t *block, uint8_t *pixels,
                                       int line_size)
 {
     x86_reg line_skip = line_size;
@@ -318,10 +318,10 @@ void ff_put_signed_pixels_clamped_mmx(const DCTELEM *block, uint8_t *pixels,
         : "memory");
 }
 
-void ff_add_pixels_clamped_mmx(const DCTELEM *block, uint8_t *pixels,
+void ff_add_pixels_clamped_mmx(const int16_t *block, uint8_t *pixels,
                                int line_size)
 {
-    const DCTELEM *p;
+    const int16_t *p;
     uint8_t *pix;
     int i;
 
@@ -423,7 +423,7 @@ static void put_pixels16_mmx(uint8_t *block, const uint8_t *pixels,
 }
 
 #define CLEAR_BLOCKS(name, n)                           \
-static void name(DCTELEM *blocks)                       \
+static void name(int16_t *blocks)                       \
 {                                                       \
     __asm__ volatile (                                  \
         "pxor %%mm7, %%mm7              \n\t"           \
@@ -443,7 +443,7 @@ static void name(DCTELEM *blocks)                       \
 CLEAR_BLOCKS(clear_blocks_mmx, 6)
 CLEAR_BLOCKS(clear_block_mmx, 1)
 
-static void clear_block_sse(DCTELEM *block)
+static void clear_block_sse(int16_t *block)
 {
     __asm__ volatile (
         "xorps  %%xmm0, %%xmm0          \n"
@@ -460,7 +460,7 @@ static void clear_block_sse(DCTELEM *block)
     );
 }
 
-static void clear_blocks_sse(DCTELEM *blocks)
+static void clear_blocks_sse(int16_t *blocks)
 {
     __asm__ volatile (
         "xorps  %%xmm0, %%xmm0              \n"
@@ -1882,28 +1882,28 @@ void ff_avg_dirac_pixels32_sse2(uint8_t *dst, const uint8_t *src[5], int stride,
  * converted. */
 #if CONFIG_GPL
 static void ff_libmpeg2mmx_idct_put(uint8_t *dest, int line_size,
-                                    DCTELEM *block)
+                                    int16_t *block)
 {
     ff_mmx_idct(block);
     ff_put_pixels_clamped_mmx(block, dest, line_size);
 }
 
 static void ff_libmpeg2mmx_idct_add(uint8_t *dest, int line_size,
-                                    DCTELEM *block)
+                                    int16_t *block)
 {
     ff_mmx_idct(block);
     ff_add_pixels_clamped_mmx(block, dest, line_size);
 }
 
 static void ff_libmpeg2mmx2_idct_put(uint8_t *dest, int line_size,
-                                     DCTELEM *block)
+                                     int16_t *block)
 {
     ff_mmxext_idct(block);
     ff_put_pixels_clamped_mmx(block, dest, line_size);
 }
 
 static void ff_libmpeg2mmx2_idct_add(uint8_t *dest, int line_size,
-                                     DCTELEM *block)
+                                     int16_t *block)
 {
     ff_mmxext_idct(block);
     ff_add_pixels_clamped_mmx(block, dest, line_size);
diff --git a/libavcodec/x86/dsputil_mmx.h b/libavcodec/x86/dsputil_mmx.h
index a0a44c1275..033a175a15 100644
--- a/libavcodec/x86/dsputil_mmx.h
+++ b/libavcodec/x86/dsputil_mmx.h
@@ -83,9 +83,9 @@ extern const double ff_pd_2[2];
 void ff_dsputilenc_init_mmx(DSPContext* c, AVCodecContext *avctx);
 void ff_dsputil_init_pix_mmx(DSPContext* c, AVCodecContext *avctx);
 
-void ff_add_pixels_clamped_mmx(const DCTELEM *block, uint8_t *pixels, int line_size);
-void ff_put_pixels_clamped_mmx(const DCTELEM *block, uint8_t *pixels, int line_size);
-void ff_put_signed_pixels_clamped_mmx(const DCTELEM *block, uint8_t *pixels, int line_size);
+void ff_add_pixels_clamped_mmx(const int16_t *block, uint8_t *pixels, int line_size);
+void ff_put_pixels_clamped_mmx(const int16_t *block, uint8_t *pixels, int line_size);
+void ff_put_signed_pixels_clamped_mmx(const int16_t *block, uint8_t *pixels, int line_size);
 
 void ff_put_cavs_qpel8_mc00_mmxext(uint8_t *dst, uint8_t *src, int stride);
 void ff_avg_cavs_qpel8_mc00_mmxext(uint8_t *dst, uint8_t *src, int stride);
@@ -100,8 +100,8 @@ void ff_put_rv40_qpel16_mc33_mmx(uint8_t *block, uint8_t *pixels, int line_size)
 void ff_avg_rv40_qpel8_mc33_mmx(uint8_t *block, uint8_t *pixels, int line_size);
 void ff_avg_rv40_qpel16_mc33_mmx(uint8_t *block, uint8_t *pixels, int line_size);
 
-void ff_mmx_idct(DCTELEM *block);
-void ff_mmxext_idct(DCTELEM *block);
+void ff_mmx_idct(int16_t *block);
+void ff_mmxext_idct(int16_t *block);
 
 
 void ff_deinterlace_line_mmx(uint8_t *dst,
diff --git a/libavcodec/x86/dsputilenc.asm b/libavcodec/x86/dsputilenc.asm
index 55eacf5b96..1839bee24a 100644
--- a/libavcodec/x86/dsputilenc.asm
+++ b/libavcodec/x86/dsputilenc.asm
@@ -335,7 +335,7 @@ cglobal sse16, 5, 5, 8
     RET
 
 INIT_MMX mmx
-; get_pixels_mmx(DCTELEM *block, const uint8_t *pixels, int line_size)
+; get_pixels_mmx(int16_t *block, const uint8_t *pixels, int line_size)
 cglobal get_pixels, 3,4
     movsxdifnidn r2, r2d
     add          r0, 128
@@ -392,7 +392,7 @@ cglobal get_pixels, 3, 4
     RET
 
 INIT_MMX mmx
-; diff_pixels_mmx(DCTELEM *block, const uint8_t *s1, const unint8_t *s2, stride)
+; diff_pixels_mmx(int16_t *block, const uint8_t *s1, const unint8_t *s2, stride)
 cglobal diff_pixels, 4,5
     movsxdifnidn r3, r3d
     pxor         m7, m7
diff --git a/libavcodec/x86/dsputilenc_mmx.c b/libavcodec/x86/dsputilenc_mmx.c
index 8a50a26ce2..5ac3c1c10a 100644
--- a/libavcodec/x86/dsputilenc_mmx.c
+++ b/libavcodec/x86/dsputilenc_mmx.c
@@ -30,9 +30,9 @@
 #include "libavcodec/mathops.h"
 #include "dsputil_mmx.h"
 
-void ff_get_pixels_mmx(DCTELEM *block, const uint8_t *pixels, int line_size);
-void ff_get_pixels_sse2(DCTELEM *block, const uint8_t *pixels, int line_size);
-void ff_diff_pixels_mmx(DCTELEM *block, const uint8_t *s1, const uint8_t *s2, int stride);
+void ff_get_pixels_mmx(int16_t *block, const uint8_t *pixels, int line_size);
+void ff_get_pixels_sse2(int16_t *block, const uint8_t *pixels, int line_size);
+void ff_diff_pixels_mmx(int16_t *block, const uint8_t *s1, const uint8_t *s2, int stride);
 int ff_pix_sum16_mmx(uint8_t * pix, int line_size);
 int ff_pix_norm1_mmx(uint8_t *pix, int line_size);
 
@@ -798,7 +798,7 @@ static void sub_hfyu_median_prediction_mmxext(uint8_t *dst, const uint8_t *src1,
     HSUM(%%xmm0, %%xmm1, %0)
 
 #define DCT_SAD_FUNC(cpu) \
-static int sum_abs_dctelem_##cpu(DCTELEM *block){\
+static int sum_abs_dctelem_##cpu(int16_t *block){\
     int sum;\
     __asm__ volatile(\
         DCT_SAD\
diff --git a/libavcodec/x86/h264_idct.asm b/libavcodec/x86/h264_idct.asm
index d0ac0fe16d..8fef7b8199 100644
--- a/libavcodec/x86/h264_idct.asm
+++ b/libavcodec/x86/h264_idct.asm
@@ -302,7 +302,7 @@ cglobal h264_idct8_dc_add_8, 3, 3, 0
 
 INIT_MMX mmx
 ; ff_h264_idct_add16_mmx(uint8_t *dst, const int *block_offset,
-;             DCTELEM *block, int stride, const uint8_t nnzc[6*8])
+;             int16_t *block, int stride, const uint8_t nnzc[6*8])
 cglobal h264_idct_add16_8, 5, 7 + npicregs, 0, dst, block_offset, block, stride, nnzc, cntr, coeff, picreg
     xor          r5, r5
 %ifdef PIC
@@ -324,7 +324,7 @@ cglobal h264_idct_add16_8, 5, 7 + npicregs, 0, dst, block_offset, block, stride,
     REP_RET
 
 ; ff_h264_idct8_add4_mmx(uint8_t *dst, const int *block_offset,
-;                        DCTELEM *block, int stride, const uint8_t nnzc[6*8])
+;                        int16_t *block, int stride, const uint8_t nnzc[6*8])
 cglobal h264_idct8_add4_8, 5, 7 + npicregs, 0, dst, block_offset, block, stride, nnzc, cntr, coeff, picreg
     %assign pad 128+4-(stack_offset&7)
     SUB         rsp, pad
@@ -357,7 +357,7 @@ cglobal h264_idct8_add4_8, 5, 7 + npicregs, 0, dst, block_offset, block, stride,
 
 INIT_MMX mmxext
 ; ff_h264_idct_add16_mmxext(uint8_t *dst, const int *block_offset,
-;                           DCTELEM *block, int stride, const uint8_t nnzc[6*8])
+;                           int16_t *block, int stride, const uint8_t nnzc[6*8])
 cglobal h264_idct_add16_8, 5, 8 + npicregs, 0, dst1, block_offset, block, stride, nnzc, cntr, coeff, dst2, picreg
     xor          r5, r5
 %ifdef PIC
@@ -402,7 +402,7 @@ cglobal h264_idct_add16_8, 5, 8 + npicregs, 0, dst1, block_offset, block, stride
 
 INIT_MMX mmx
 ; ff_h264_idct_add16intra_mmx(uint8_t *dst, const int *block_offset,
-;                             DCTELEM *block, int stride, const uint8_t nnzc[6*8])
+;                             int16_t *block, int stride, const uint8_t nnzc[6*8])
 cglobal h264_idct_add16intra_8, 5, 7 + npicregs, 0, dst, block_offset, block, stride, nnzc, cntr, coeff, picreg
     xor          r5, r5
 %ifdef PIC
@@ -426,7 +426,7 @@ cglobal h264_idct_add16intra_8, 5, 7 + npicregs, 0, dst, block_offset, block, st
 
 INIT_MMX mmxext
 ; ff_h264_idct_add16intra_mmxext(uint8_t *dst, const int *block_offset,
-;                                DCTELEM *block, int stride,
+;                                int16_t *block, int stride,
 ;                                const uint8_t nnzc[6*8])
 cglobal h264_idct_add16intra_8, 5, 8 + npicregs, 0, dst1, block_offset, block, stride, nnzc, cntr, coeff, dst2, picreg
     xor          r5, r5
@@ -469,7 +469,7 @@ cglobal h264_idct_add16intra_8, 5, 8 + npicregs, 0, dst1, block_offset, block, s
     REP_RET
 
 ; ff_h264_idct8_add4_mmxext(uint8_t *dst, const int *block_offset,
-;                           DCTELEM *block, int stride,
+;                           int16_t *block, int stride,
 ;                           const uint8_t nnzc[6*8])
 cglobal h264_idct8_add4_8, 5, 8 + npicregs, 0, dst1, block_offset, block, stride, nnzc, cntr, coeff, dst2, picreg
     %assign pad 128+4-(stack_offset&7)
@@ -530,7 +530,7 @@ cglobal h264_idct8_add4_8, 5, 8 + npicregs, 0, dst1, block_offset, block, stride
 
 INIT_XMM sse2
 ; ff_h264_idct8_add4_sse2(uint8_t *dst, const int *block_offset,
-;                         DCTELEM *block, int stride, const uint8_t nnzc[6*8])
+;                         int16_t *block, int stride, const uint8_t nnzc[6*8])
 cglobal h264_idct8_add4_8, 5, 8 + npicregs, 10, dst1, block_offset, block, stride, nnzc, cntr, coeff, dst2, picreg
     xor          r5, r5
 %ifdef PIC
@@ -605,7 +605,7 @@ h264_idct_add8_mmx_plane:
     rep ret
 
 ; ff_h264_idct_add8_mmx(uint8_t **dest, const int *block_offset,
-;                       DCTELEM *block, int stride, const uint8_t nnzc[6*8])
+;                       int16_t *block, int stride, const uint8_t nnzc[6*8])
 cglobal h264_idct_add8_8, 5, 8 + npicregs, 0, dst1, block_offset, block, stride, nnzc, cntr, coeff, dst2, picreg
     mov          r5, 16
     add          r2, 512
@@ -669,7 +669,7 @@ h264_idct_add8_mmxext_plane:
 
 INIT_MMX mmxext
 ; ff_h264_idct_add8_mmxext(uint8_t **dest, const int *block_offset,
-;                          DCTELEM *block, int stride, const uint8_t nnzc[6*8])
+;                          int16_t *block, int stride, const uint8_t nnzc[6*8])
 cglobal h264_idct_add8_8, 5, 8 + npicregs, 0, dst1, block_offset, block, stride, nnzc, cntr, coeff, dst2, picreg
     mov          r5, 16
     add          r2, 512
@@ -746,7 +746,7 @@ h264_add8x4_idct_sse2:
 %endmacro
 
 ; ff_h264_idct_add16_sse2(uint8_t *dst, const int *block_offset,
-;                         DCTELEM *block, int stride, const uint8_t nnzc[6*8])
+;                         int16_t *block, int stride, const uint8_t nnzc[6*8])
 cglobal h264_idct_add16_8, 5, 5 + ARCH_X86_64, 8
 %if ARCH_X86_64
     mov         r5, r0
@@ -793,7 +793,7 @@ cglobal h264_idct_add16_8, 5, 5 + ARCH_X86_64, 8
 %endmacro
 
 ; ff_h264_idct_add16intra_sse2(uint8_t *dst, const int *block_offset,
-;                              DCTELEM *block, int stride, const uint8_t nnzc[6*8])
+;                              int16_t *block, int stride, const uint8_t nnzc[6*8])
 cglobal h264_idct_add16intra_8, 5, 7 + ARCH_X86_64, 8
 %if ARCH_X86_64
     mov         r7, r0
@@ -844,7 +844,7 @@ cglobal h264_idct_add16intra_8, 5, 7 + ARCH_X86_64, 8
 %endmacro
 
 ; ff_h264_idct_add8_sse2(uint8_t **dest, const int *block_offset,
-;                        DCTELEM *block, int stride, const uint8_t nnzc[6*8])
+;                        int16_t *block, int stride, const uint8_t nnzc[6*8])
 cglobal h264_idct_add8_8, 5, 7 + ARCH_X86_64, 8
     add          r2, 512
 %if ARCH_X86_64
@@ -861,7 +861,7 @@ cglobal h264_idct_add8_8, 5, 7 + ARCH_X86_64, 8
     add8_sse2_cycle 3, 0x64
     RET
 
-;void ff_h264_luma_dc_dequant_idct_mmx(DCTELEM *output, DCTELEM *input, int qmul)
+;void ff_h264_luma_dc_dequant_idct_mmx(int16_t *output, int16_t *input, int qmul)
 
 %macro WALSH4_1D 5
     SUMSUB_BADC w, %4, %3, %2, %1, %5
diff --git a/libavcodec/x86/h264_intrapred_init.c b/libavcodec/x86/h264_intrapred_init.c
index deededaddb..015cc6ff19 100644
--- a/libavcodec/x86/h264_intrapred_init.c
+++ b/libavcodec/x86/h264_intrapred_init.c
@@ -20,6 +20,7 @@
 
 #include "libavutil/cpu.h"
 #include "libavutil/x86/cpu.h"
+#include "libavcodec/avcodec.h"
 #include "libavcodec/h264pred.h"
 
 #define PRED4x4(TYPE, DEPTH, OPT) \
diff --git a/libavcodec/x86/h264dsp_init.c b/libavcodec/x86/h264dsp_init.c
index 65d300fe4d..a0c22eb186 100644
--- a/libavcodec/x86/h264dsp_init.c
+++ b/libavcodec/x86/h264dsp_init.c
@@ -48,7 +48,7 @@ IDCT_ADD_FUNC(8, 10, avx)
 #define IDCT_ADD_REP_FUNC(NUM, REP, DEPTH, OPT)                         \
 void ff_h264_idct ## NUM ## _add ## REP ## _ ## DEPTH ## _ ## OPT       \
     (uint8_t *dst, const int *block_offset,                             \
-     DCTELEM *block, int stride, const uint8_t nnzc[6 * 8]);
+     int16_t *block, int stride, const uint8_t nnzc[6 * 8]);
 
 IDCT_ADD_REP_FUNC(8, 4, 8, mmx)
 IDCT_ADD_REP_FUNC(8, 4, 8, mmxext)
@@ -70,7 +70,7 @@ IDCT_ADD_REP_FUNC(, 16intra, 10, avx)
 #define IDCT_ADD_REP_FUNC2(NUM, REP, DEPTH, OPT)                      \
 void ff_h264_idct ## NUM ## _add ## REP ## _ ## DEPTH ## _ ## OPT     \
     (uint8_t **dst, const int *block_offset,                          \
-     DCTELEM *block, int stride, const uint8_t nnzc[6 * 8]);
+     int16_t *block, int stride, const uint8_t nnzc[6 * 8]);
 
 IDCT_ADD_REP_FUNC2(, 8, 8, mmx)
 IDCT_ADD_REP_FUNC2(, 8, 8, mmxext)
@@ -78,8 +78,8 @@ IDCT_ADD_REP_FUNC2(, 8, 8, sse2)
 IDCT_ADD_REP_FUNC2(, 8, 10, sse2)
 IDCT_ADD_REP_FUNC2(, 8, 10, avx)
 
-void ff_h264_luma_dc_dequant_idct_mmx(DCTELEM *output, DCTELEM *input, int qmul);
-void ff_h264_luma_dc_dequant_idct_sse2(DCTELEM *output, DCTELEM *input, int qmul);
+void ff_h264_luma_dc_dequant_idct_mmx(int16_t *output, int16_t *input, int qmul);
+void ff_h264_luma_dc_dequant_idct_sse2(int16_t *output, int16_t *input, int qmul);
 
 /***********************************/
 /* deblocking */
diff --git a/libavcodec/x86/idct_mmx_xvid.c b/libavcodec/x86/idct_mmx_xvid.c
index b62865e679..5e9f405452 100644
--- a/libavcodec/x86/idct_mmx_xvid.c
+++ b/libavcodec/x86/idct_mmx_xvid.c
@@ -531,25 +531,25 @@ __asm__ volatile(
     :: "r"(block), "r"(rounder_0), "r"(tab_i_04_xmm), "r"(tg_1_16));
 }
 
-void ff_idct_xvid_mmx_put(uint8_t *dest, int line_size, DCTELEM *block)
+void ff_idct_xvid_mmx_put(uint8_t *dest, int line_size, int16_t *block)
 {
     ff_idct_xvid_mmx(block);
     ff_put_pixels_clamped_mmx(block, dest, line_size);
 }
 
-void ff_idct_xvid_mmx_add(uint8_t *dest, int line_size, DCTELEM *block)
+void ff_idct_xvid_mmx_add(uint8_t *dest, int line_size, int16_t *block)
 {
     ff_idct_xvid_mmx(block);
     ff_add_pixels_clamped_mmx(block, dest, line_size);
 }
 
-void ff_idct_xvid_mmxext_put(uint8_t *dest, int line_size, DCTELEM *block)
+void ff_idct_xvid_mmxext_put(uint8_t *dest, int line_size, int16_t *block)
 {
     ff_idct_xvid_mmxext(block);
     ff_put_pixels_clamped_mmx(block, dest, line_size);
 }
 
-void ff_idct_xvid_mmxext_add(uint8_t *dest, int line_size, DCTELEM *block)
+void ff_idct_xvid_mmxext_add(uint8_t *dest, int line_size, int16_t *block)
 {
     ff_idct_xvid_mmxext(block);
     ff_add_pixels_clamped_mmx(block, dest, line_size);
diff --git a/libavcodec/x86/idct_xvid.h b/libavcodec/x86/idct_xvid.h
index a584fba377..7a2847b864 100644
--- a/libavcodec/x86/idct_xvid.h
+++ b/libavcodec/x86/idct_xvid.h
@@ -28,15 +28,13 @@
 
 #include <stdint.h>
 
-#include "libavcodec/dsputil.h"
-
 void ff_idct_xvid_mmx(short *block);
-void ff_idct_xvid_mmx_put(uint8_t *dest, int line_size, DCTELEM *block);
-void ff_idct_xvid_mmx_add(uint8_t *dest, int line_size, DCTELEM *block);
+void ff_idct_xvid_mmx_put(uint8_t *dest, int line_size, int16_t *block);
+void ff_idct_xvid_mmx_add(uint8_t *dest, int line_size, int16_t *block);
 
 void ff_idct_xvid_mmxext(short *block);
-void ff_idct_xvid_mmxext_put(uint8_t *dest, int line_size, DCTELEM *block);
-void ff_idct_xvid_mmxext_add(uint8_t *dest, int line_size, DCTELEM *block);
+void ff_idct_xvid_mmxext_put(uint8_t *dest, int line_size, int16_t *block);
+void ff_idct_xvid_mmxext_add(uint8_t *dest, int line_size, int16_t *block);
 
 void ff_idct_xvid_sse2(short *block);
 void ff_idct_xvid_sse2_put(uint8_t *dest, int line_size, short *block);
diff --git a/libavcodec/x86/mpegvideo.c b/libavcodec/x86/mpegvideo.c
index 9870ed7d51..f4bf05ffdd 100644
--- a/libavcodec/x86/mpegvideo.c
+++ b/libavcodec/x86/mpegvideo.c
@@ -29,7 +29,7 @@
 #if HAVE_INLINE_ASM
 
 static void dct_unquantize_h263_intra_mmx(MpegEncContext *s,
-                                  DCTELEM *block, int n, int qscale)
+                                  int16_t *block, int n, int qscale)
 {
     x86_reg level, qmul, qadd, nCoeffs;
 
@@ -104,7 +104,7 @@ __asm__ volatile(
 
 
 static void dct_unquantize_h263_inter_mmx(MpegEncContext *s,
-                                  DCTELEM *block, int n, int qscale)
+                                  int16_t *block, int n, int qscale)
 {
     x86_reg qmul, qadd, nCoeffs;
 
@@ -187,7 +187,7 @@ __asm__ volatile(
  high3 += tlow1
 */
 static void dct_unquantize_mpeg1_intra_mmx(MpegEncContext *s,
-                                     DCTELEM *block, int n, int qscale)
+                                     int16_t *block, int n, int qscale)
 {
     x86_reg nCoeffs;
     const uint16_t *quant_matrix;
@@ -256,7 +256,7 @@ __asm__ volatile(
 }
 
 static void dct_unquantize_mpeg1_inter_mmx(MpegEncContext *s,
-                                     DCTELEM *block, int n, int qscale)
+                                     int16_t *block, int n, int qscale)
 {
     x86_reg nCoeffs;
     const uint16_t *quant_matrix;
@@ -322,7 +322,7 @@ __asm__ volatile(
 }
 
 static void dct_unquantize_mpeg2_intra_mmx(MpegEncContext *s,
-                                     DCTELEM *block, int n, int qscale)
+                                     int16_t *block, int n, int qscale)
 {
     x86_reg nCoeffs;
     const uint16_t *quant_matrix;
@@ -388,7 +388,7 @@ __asm__ volatile(
 }
 
 static void dct_unquantize_mpeg2_inter_mmx(MpegEncContext *s,
-                                     DCTELEM *block, int n, int qscale)
+                                     int16_t *block, int n, int qscale)
 {
     x86_reg nCoeffs;
     const uint16_t *quant_matrix;
@@ -464,7 +464,7 @@ __asm__ volatile(
         );
 }
 
-static void  denoise_dct_mmx(MpegEncContext *s, DCTELEM *block){
+static void  denoise_dct_mmx(MpegEncContext *s, int16_t *block){
     const int intra= s->mb_intra;
     int *sum= s->dct_error_sum[intra];
     uint16_t *offset= s->dct_offset[intra];
@@ -518,7 +518,7 @@ static void  denoise_dct_mmx(MpegEncContext *s, DCTELEM *block){
     );
 }
 
-static void  denoise_dct_sse2(MpegEncContext *s, DCTELEM *block){
+static void  denoise_dct_sse2(MpegEncContext *s, int16_t *block){
     const int intra= s->mb_intra;
     int *sum= s->dct_error_sum[intra];
     uint16_t *offset= s->dct_offset[intra];
diff --git a/libavcodec/x86/mpegvideoenc_template.c b/libavcodec/x86/mpegvideoenc_template.c
index 47c3b43f94..1e0505ea3c 100644
--- a/libavcodec/x86/mpegvideoenc_template.c
+++ b/libavcodec/x86/mpegvideoenc_template.c
@@ -92,7 +92,7 @@
 #endif
 
 static int RENAME(dct_quantize)(MpegEncContext *s,
-                            DCTELEM *block, int n,
+                            int16_t *block, int n,
                             int qscale, int *overflow)
 {
     x86_reg last_non_zero_p1;
diff --git a/libavcodec/x86/proresdsp.asm b/libavcodec/x86/proresdsp.asm
index 094f872bc8..aedacc2cdc 100644
--- a/libavcodec/x86/proresdsp.asm
+++ b/libavcodec/x86/proresdsp.asm
@@ -232,7 +232,7 @@ section .text align=16
 %endmacro
 
 ; void prores_idct_put_10_<opt>(uint8_t *pixels, int stride,
-;                               DCTELEM *block, const int16_t *qmat);
+;                               int16_t *block, const int16_t *qmat);
 %macro idct_put_fn 1
 cglobal prores_idct_put_10, 4, 4, %1
     movsxd      r1,  r1d
diff --git a/libavcodec/x86/proresdsp_init.c b/libavcodec/x86/proresdsp_init.c
index 21ce0986d2..91ff257564 100644
--- a/libavcodec/x86/proresdsp_init.c
+++ b/libavcodec/x86/proresdsp_init.c
@@ -21,14 +21,15 @@
  */
 
 #include "libavutil/x86/cpu.h"
+#include "libavcodec/dsputil.h"
 #include "libavcodec/proresdsp.h"
 
 void ff_prores_idct_put_10_sse2(uint16_t *dst, int linesize,
-                                DCTELEM *block, const int16_t *qmat);
+                                int16_t *block, const int16_t *qmat);
 void ff_prores_idct_put_10_sse4(uint16_t *dst, int linesize,
-                                DCTELEM *block, const int16_t *qmat);
+                                int16_t *block, const int16_t *qmat);
 void ff_prores_idct_put_10_avx (uint16_t *dst, int linesize,
-                                DCTELEM *block, const int16_t *qmat);
+                                int16_t *block, const int16_t *qmat);
 
 void ff_proresdsp_x86_init(ProresDSPContext *dsp, AVCodecContext *avctx)
 {
diff --git a/libavcodec/x86/rv34dsp.asm b/libavcodec/x86/rv34dsp.asm
index c099ac5b1f..4d9c35b600 100644
--- a/libavcodec/x86/rv34dsp.asm
+++ b/libavcodec/x86/rv34dsp.asm
@@ -133,7 +133,7 @@ cglobal rv34_idct_dc_add, 3, 3
     mova        mm5, [pd_512]           ; 0x200
 %endmacro
 
-; ff_rv34_idct_add_mmxext(uint8_t *dst, ptrdiff_t stride, DCTELEM *block);
+; ff_rv34_idct_add_mmxext(uint8_t *dst, ptrdiff_t stride, int16_t *block);
 %macro COL_TRANSFORM  4
     pshufw      mm3, %2, 0xDD        ; col. 1,3,1,3
     pshufw       %2, %2, 0x88        ; col. 0,2,0,2
diff --git a/libavcodec/x86/rv34dsp_init.c b/libavcodec/x86/rv34dsp_init.c
index 6b6cf914ea..a93096791b 100644
--- a/libavcodec/x86/rv34dsp_init.c
+++ b/libavcodec/x86/rv34dsp_init.c
@@ -25,11 +25,11 @@
 #include "libavcodec/dsputil.h"
 #include "libavcodec/rv34dsp.h"
 
-void ff_rv34_idct_dc_mmxext(DCTELEM *block);
-void ff_rv34_idct_dc_noround_mmxext(DCTELEM *block);
+void ff_rv34_idct_dc_mmxext(int16_t *block);
+void ff_rv34_idct_dc_noround_mmxext(int16_t *block);
 void ff_rv34_idct_dc_add_mmx(uint8_t *dst, ptrdiff_t stride, int dc);
 void ff_rv34_idct_dc_add_sse4(uint8_t *dst, ptrdiff_t stride, int dc);
-void ff_rv34_idct_add_mmxext(uint8_t *dst, ptrdiff_t stride, DCTELEM *block);
+void ff_rv34_idct_add_mmxext(uint8_t *dst, ptrdiff_t stride, int16_t *block);
 
 av_cold void ff_rv34dsp_init_x86(RV34DSPContext* c, DSPContext *dsp)
 {
diff --git a/libavcodec/x86/simple_idct.c b/libavcodec/x86/simple_idct.c
index c514d755ee..f9fd7dc3c9 100644
--- a/libavcodec/x86/simple_idct.c
+++ b/libavcodec/x86/simple_idct.c
@@ -1154,12 +1154,12 @@ void ff_simple_idct_mmx(int16_t *block)
 
 //FIXME merge add/put into the idct
 
-void ff_simple_idct_put_mmx(uint8_t *dest, int line_size, DCTELEM *block)
+void ff_simple_idct_put_mmx(uint8_t *dest, int line_size, int16_t *block)
 {
     idct(block);
     ff_put_pixels_clamped_mmx(block, dest, line_size);
 }
-void ff_simple_idct_add_mmx(uint8_t *dest, int line_size, DCTELEM *block)
+void ff_simple_idct_add_mmx(uint8_t *dest, int line_size, int16_t *block)
 {
     idct(block);
     ff_add_pixels_clamped_mmx(block, dest, line_size);
diff --git a/libavcodec/x86/vc1dsp_mmx.c b/libavcodec/x86/vc1dsp_mmx.c
index 6ce65e4e16..a7695357d9 100644
--- a/libavcodec/x86/vc1dsp_mmx.c
+++ b/libavcodec/x86/vc1dsp_mmx.c
@@ -493,7 +493,7 @@ DECLARE_FUNCTION(3, 2)
 DECLARE_FUNCTION(3, 3)
 
 static void vc1_inv_trans_4x4_dc_mmxext(uint8_t *dest, int linesize,
-                                        DCTELEM *block)
+                                        int16_t *block)
 {
     int dc = block[0];
     dc = (17 * dc +  4) >> 3;
@@ -532,7 +532,7 @@ static void vc1_inv_trans_4x4_dc_mmxext(uint8_t *dest, int linesize,
 }
 
 static void vc1_inv_trans_4x8_dc_mmxext(uint8_t *dest, int linesize,
-                                        DCTELEM *block)
+                                        int16_t *block)
 {
     int dc = block[0];
     dc = (17 * dc +  4) >> 3;
@@ -594,7 +594,7 @@ static void vc1_inv_trans_4x8_dc_mmxext(uint8_t *dest, int linesize,
 }
 
 static void vc1_inv_trans_8x4_dc_mmxext(uint8_t *dest, int linesize,
-                                        DCTELEM *block)
+                                        int16_t *block)
 {
     int dc = block[0];
     dc = ( 3 * dc +  1) >> 1;
@@ -633,7 +633,7 @@ static void vc1_inv_trans_8x4_dc_mmxext(uint8_t *dest, int linesize,
 }
 
 static void vc1_inv_trans_8x8_dc_mmxext(uint8_t *dest, int linesize,
-                                        DCTELEM *block)
+                                        int16_t *block)
 {
     int dc = block[0];
     dc = (3 * dc +  1) >> 1;
diff --git a/libavcodec/x86/vp3dsp_init.c b/libavcodec/x86/vp3dsp_init.c
index e425f503ac..1df570ec0b 100644
--- a/libavcodec/x86/vp3dsp_init.c
+++ b/libavcodec/x86/vp3dsp_init.c
@@ -25,17 +25,18 @@
 #include "libavutil/x86/cpu.h"
 #include "libavutil/x86/asm.h"
 #include "libavcodec/avcodec.h"
+#include "libavcodec/dsputil.h"
 #include "libavcodec/vp3dsp.h"
 #include "config.h"
 
-void ff_vp3_idct_put_mmx(uint8_t *dest, int line_size, DCTELEM *block);
-void ff_vp3_idct_add_mmx(uint8_t *dest, int line_size, DCTELEM *block);
+void ff_vp3_idct_put_mmx(uint8_t *dest, int line_size, int16_t *block);
+void ff_vp3_idct_add_mmx(uint8_t *dest, int line_size, int16_t *block);
 
-void ff_vp3_idct_put_sse2(uint8_t *dest, int line_size, DCTELEM *block);
-void ff_vp3_idct_add_sse2(uint8_t *dest, int line_size, DCTELEM *block);
+void ff_vp3_idct_put_sse2(uint8_t *dest, int line_size, int16_t *block);
+void ff_vp3_idct_add_sse2(uint8_t *dest, int line_size, int16_t *block);
 
 void ff_vp3_idct_dc_add_mmxext(uint8_t *dest, int line_size,
-                               DCTELEM *block);
+                               int16_t *block);
 
 void ff_vp3_v_loop_filter_mmxext(uint8_t *src, int stride,
                                  int *bounding_values);
diff --git a/libavcodec/x86/vp8dsp.asm b/libavcodec/x86/vp8dsp.asm
index 84805fdc66..c5fb4555d3 100644
--- a/libavcodec/x86/vp8dsp.asm
+++ b/libavcodec/x86/vp8dsp.asm
@@ -899,7 +899,7 @@ cglobal put_vp8_pixels16, 5, 5, 2, dst, dststride, src, srcstride, height
     REP_RET
 
 ;-----------------------------------------------------------------------------
-; void vp8_idct_dc_add_<opt>(uint8_t *dst, DCTELEM block[16], int stride);
+; void vp8_idct_dc_add_<opt>(uint8_t *dst, int16_t block[16], int stride);
 ;-----------------------------------------------------------------------------
 
 %macro ADD_DC 4
@@ -977,7 +977,7 @@ cglobal vp8_idct_dc_add, 3, 3, 6, dst, block, stride
     RET
 
 ;-----------------------------------------------------------------------------
-; void vp8_idct_dc_add4y_<opt>(uint8_t *dst, DCTELEM block[4][16], int stride);
+; void vp8_idct_dc_add4y_<opt>(uint8_t *dst, int16_t block[4][16], int stride);
 ;-----------------------------------------------------------------------------
 
 %if ARCH_X86_32
@@ -1050,7 +1050,7 @@ cglobal vp8_idct_dc_add4y, 3, 3, 6, dst, block, stride
     RET
 
 ;-----------------------------------------------------------------------------
-; void vp8_idct_dc_add4uv_<opt>(uint8_t *dst, DCTELEM block[4][16], int stride);
+; void vp8_idct_dc_add4uv_<opt>(uint8_t *dst, int16_t block[4][16], int stride);
 ;-----------------------------------------------------------------------------
 
 INIT_MMX mmx
@@ -1092,7 +1092,7 @@ cglobal vp8_idct_dc_add4uv, 3, 3, 0, dst, block, stride
     RET
 
 ;-----------------------------------------------------------------------------
-; void vp8_idct_add_<opt>(uint8_t *dst, DCTELEM block[16], int stride);
+; void vp8_idct_add_<opt>(uint8_t *dst, int16_t block[16], int stride);
 ;-----------------------------------------------------------------------------
 
 ; calculate %1=mul_35468(%1)-mul_20091(%2); %2=mul_20091(%1)+mul_35468(%2)
@@ -1172,7 +1172,7 @@ INIT_MMX sse
 VP8_IDCT_ADD
 
 ;-----------------------------------------------------------------------------
-; void vp8_luma_dc_wht_mmxext(DCTELEM block[4][4][16], DCTELEM dc[16])
+; void vp8_luma_dc_wht_mmxext(int16_t block[4][4][16], int16_t dc[16])
 ;-----------------------------------------------------------------------------
 
 %macro SCATTER_WHT 3
diff --git a/libavcodec/x86/vp8dsp_init.c b/libavcodec/x86/vp8dsp_init.c
index 735619c32b..09e2d9109e 100644
--- a/libavcodec/x86/vp8dsp_init.c
+++ b/libavcodec/x86/vp8dsp_init.c
@@ -230,21 +230,21 @@ 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],
+extern void ff_vp8_idct_dc_add_mmx(uint8_t *dst, int16_t block[16],
                                    ptrdiff_t stride);
-extern void ff_vp8_idct_dc_add_sse4(uint8_t *dst, DCTELEM block[16],
+extern void ff_vp8_idct_dc_add_sse4(uint8_t *dst, int16_t block[16],
                                     ptrdiff_t stride);
-extern void ff_vp8_idct_dc_add4y_mmx(uint8_t *dst, DCTELEM block[4][16],
+extern void ff_vp8_idct_dc_add4y_mmx(uint8_t *dst, int16_t block[4][16],
                                       ptrdiff_t stride);
-extern void ff_vp8_idct_dc_add4y_sse2(uint8_t *dst, DCTELEM block[4][16],
+extern void ff_vp8_idct_dc_add4y_sse2(uint8_t *dst, int16_t block[4][16],
                                       ptrdiff_t stride);
-extern void ff_vp8_idct_dc_add4uv_mmx(uint8_t *dst, DCTELEM block[2][16],
+extern void ff_vp8_idct_dc_add4uv_mmx(uint8_t *dst, int16_t 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],
+extern void ff_vp8_luma_dc_wht_mmx(int16_t block[4][4][16], int16_t dc[16]);
+extern void ff_vp8_luma_dc_wht_sse(int16_t block[4][4][16], int16_t dc[16]);
+extern void ff_vp8_idct_add_mmx(uint8_t *dst, int16_t block[16],
                                 ptrdiff_t stride);
-extern void ff_vp8_idct_add_sse(uint8_t *dst, DCTELEM block[16],
+extern void ff_vp8_idct_add_sse(uint8_t *dst, int16_t block[16],
                                 ptrdiff_t stride);
 
 #define DECLARE_LOOP_FILTER(NAME)\