comparison x86/dsputil_mmx.c @ 12436:d6d0a43848b4 libavcodec

Move VP3 IDCT functions from inline ASM to YASM. This fixes part of the VP3/5/6 issues on Win64.

12435:fe78a4548d12

#include "libavcodec/dsputil.h"
#include "libavcodec/h264dsp.h"
#include "libavcodec/mpegvideo.h"
#include "libavcodec/simple_idct.h"
#include "dsputil_mmx.h"
#include "vp3dsp_mmx.h"
#include "vp3dsp_sse2.h"
#include "idct_xvid.h"

//#undef NDEBUG
//#include <assert.h>

        " js 1b                             \n\t"
        :"+r"(reglen), "+r"(dst), "+r"(src)
    );
}

void ff_float_to_int16_interleave6_sse(int16_t *dst, const float **src, int len);
void ff_float_to_int16_interleave6_3dnow(int16_t *dst, const float **src, int len);
void ff_float_to_int16_interleave6_3dn2(int16_t *dst, const float **src, int len);
int32_t ff_scalarproduct_int16_mmx2(const int16_t *v1, const int16_t *v2, int order, int shift);
int32_t ff_scalarproduct_int16_sse2(const int16_t *v1, const int16_t *v2, int order, int shift);

12436:d6d0a43848b4

#include "libavcodec/dsputil.h"
#include "libavcodec/h264dsp.h"
#include "libavcodec/mpegvideo.h"
#include "libavcodec/simple_idct.h"
#include "dsputil_mmx.h"
#include "idct_xvid.h"

//#undef NDEBUG
//#include <assert.h>

        " js 1b                             \n\t"
        :"+r"(reglen), "+r"(dst), "+r"(src)
    );
}

void ff_vp3_idct_mmx(int16_t *input_data);
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_dc_add_mmx2(uint8_t *dest, int line_size, const DCTELEM *block);

void ff_vp3_v_loop_filter_mmx2(uint8_t *src, int stride, int *bounding_values);
void ff_vp3_h_loop_filter_mmx2(uint8_t *src, int stride, int *bounding_values);

void ff_vp3_idct_sse2(int16_t *input_data);
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_float_to_int16_interleave6_sse(int16_t *dst, const float **src, int len);
void ff_float_to_int16_interleave6_3dnow(int16_t *dst, const float **src, int len);
void ff_float_to_int16_interleave6_3dn2(int16_t *dst, const float **src, int len);
int32_t ff_scalarproduct_int16_mmx2(const int16_t *v1, const int16_t *v2, int order, int shift);
int32_t ff_scalarproduct_int16_sse2(const int16_t *v1, const int16_t *v2, int order, int shift);