* Copyright (c) 2000, 2001, 2002 Fabrice Bellard.
* Copyright (c) 2002-2004 Michael Niedermayer <michaelni@gmx.at>
*
- * This library is free software; you can redistribute it and/or
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
- * version 2 of the License, or (at your option) any later version.
+ * version 2.1 of the License, or (at your option) any later version.
*
- * This library is distributed in the hope that it will be useful,
+ * FFmpeg is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with this library; if not, write to the Free Software
+ * License along with FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
void ff_h264_lowres_idct_add_c(uint8_t *dst, int stride, DCTELEM *block);
void ff_h264_lowres_idct_put_c(uint8_t *dst, int stride, DCTELEM *block);
+void ff_vector_fmul_add_add_c(float *dst, const float *src0, const float *src1,
+ const float *src2, int src3, int blocksize, int step);
+void ff_float_to_int16_c(int16_t *dst, const float *src, int len);
+
/* encoding scans */
extern const uint8_t ff_alternate_horizontal_scan[64];
extern const uint8_t ff_alternate_vertical_scan[64];
#define MAX_NEG_CROP 1024
/* temporary */
-extern uint32_t squareTbl[512];
-extern uint8_t cropTbl[256 + 2 * MAX_NEG_CROP];
+extern uint32_t ff_squareTbl[512];
+extern uint8_t ff_cropTbl[256 + 2 * MAX_NEG_CROP];
/* VP3 DSP functions */
void ff_vp3_idct_c(DCTELEM *block/* align 16*/);
void ff_shrink44(uint8_t *dst, int dst_wrap, const uint8_t *src, int src_wrap, int width, int height);
void ff_shrink88(uint8_t *dst, int dst_wrap, const uint8_t *src, int src_wrap, int width, int height);
+void ff_gmc_c(uint8_t *dst, uint8_t *src, int stride, int h, int ox, int oy,
+ int dxx, int dxy, int dyx, int dyy, int shift, int r, int width, int height);
+
/* minimum alignment rules ;)
if u notice errors in the align stuff, need more alignment for some asm code for some cpu
or need to use a function with less aligned data then send a mail to the ffmpeg-dev list, ...
* h264 Chram MC
*/
h264_chroma_mc_func put_h264_chroma_pixels_tab[3];
+ /* This is really one func used in VC-1 decoding */
+ h264_chroma_mc_func put_no_rnd_h264_chroma_pixels_tab[3];
h264_chroma_mc_func avg_h264_chroma_pixels_tab[3];
qpel_mc_func put_h264_qpel_pixels_tab[4][16];
qpel_mc_func avg_h264_qpel_pixels_tab[4][16];
+ qpel_mc_func put_2tap_qpel_pixels_tab[4][16];
+ qpel_mc_func avg_2tap_qpel_pixels_tab[4][16];
+
h264_weight_func weight_h264_pixels_tab[10];
h264_biweight_func biweight_h264_pixels_tab[10];
+ /* AVS specific */
+ qpel_mc_func put_cavs_qpel_pixels_tab[2][16];
+ qpel_mc_func avg_cavs_qpel_pixels_tab[2][16];
+ void (*cavs_filter_lv)(uint8_t *pix, int stride, int alpha, int beta, int tc, int bs1, int bs2);
+ void (*cavs_filter_lh)(uint8_t *pix, int stride, int alpha, int beta, int tc, int bs1, int bs2);
+ void (*cavs_filter_cv)(uint8_t *pix, int stride, int alpha, int beta, int tc, int bs1, int bs2);
+ void (*cavs_filter_ch)(uint8_t *pix, int stride, int alpha, int beta, int tc, int bs1, int bs2);
+ void (*cavs_idct8_add)(uint8_t *dst, DCTELEM *block, int stride);
+
me_cmp_func pix_abs[2][4];
/* huffyuv specific */
void (*h264_h_loop_filter_chroma)(uint8_t *pix, int stride, int alpha, int beta, int8_t *tc0);
void (*h264_v_loop_filter_chroma_intra)(uint8_t *pix, int stride, int alpha, int beta);
void (*h264_h_loop_filter_chroma_intra)(uint8_t *pix, int stride, int alpha, int beta);
+ // h264_loop_filter_strength: simd only. the C version is inlined in h264.c
+ void (*h264_loop_filter_strength)(int16_t bS[2][4][4], uint8_t nnz[40], int8_t ref[2][40], int16_t mv[2][40][2],
+ int bidir, int edges, int step, int mask_mv0, int mask_mv1);
void (*h263_v_loop_filter)(uint8_t *src, int stride, int qscale);
void (*h263_h_loop_filter)(uint8_t *src, int stride, int qscale);
void (*h261_loop_filter)(uint8_t *src, int stride);
+ /* assume len is a multiple of 4, and arrays are 16-byte aligned */
+ void (*vorbis_inverse_coupling)(float *mag, float *ang, int blocksize);
+ /* assume len is a multiple of 8, and arrays are 16-byte aligned */
+ void (*vector_fmul)(float *dst, const float *src, int len);
+ void (*vector_fmul_reverse)(float *dst, const float *src0, const float *src1, int len);
+ /* assume len is a multiple of 8, and src arrays are 16-byte aligned */
+ void (*vector_fmul_add_add)(float *dst, const float *src0, const float *src1, const float *src2, int src3, int len, int step);
+
+ /* C version: convert floats from the range [384.0,386.0] to ints in [-32768,32767]
+ * simd versions: convert floats from [-32768.0,32767.0] without rescaling and arrays are 16byte aligned */
+ void (*float_to_int16)(int16_t *dst, const float *src, int len);
+
/* (I)DCT */
void (*fdct)(DCTELEM *block/* align 16*/);
void (*fdct248)(DCTELEM *block/* align 16*/);
#define BASIS_SHIFT 16
#define RECON_SHIFT 6
+ /* h264 functions */
void (*h264_idct_add)(uint8_t *dst, DCTELEM *block, int stride);
void (*h264_idct8_add)(uint8_t *dst, DCTELEM *block, int stride);
void (*h264_idct_dc_add)(uint8_t *dst, DCTELEM *block, int stride);
void (*h264_idct8_dc_add)(uint8_t *dst, DCTELEM *block, int stride);
+ void (*h264_dct)(DCTELEM block[4][4]);
/* snow wavelet */
void (*vertical_compose97i)(DWTELEM *b0, DWTELEM *b1, DWTELEM *b2, DWTELEM *b3, DWTELEM *b4, DWTELEM *b5, int width);
void (*horizontal_compose97i)(DWTELEM *b, int width);
- void (*inner_add_yblock)(uint8_t *obmc, const int obmc_stride, uint8_t * * block, int b_w, int b_h, int src_x, int src_y, int src_stride, slice_buffer * sb, int add, uint8_t * dst8);
+ void (*inner_add_yblock)(const uint8_t *obmc, const int obmc_stride, uint8_t * * block, int b_w, int b_h, int src_x, int src_y, int src_stride, slice_buffer * sb, int add, uint8_t * dst8);
void (*prefetch)(void *mem, int stride, int h);
void (*shrink[4])(uint8_t *dst, int dst_wrap, const uint8_t *src, int src_wrap, int width, int height);
+
+ /* vc1 functions */
+ void (*vc1_inv_trans_8x8)(DCTELEM *b);
+ void (*vc1_inv_trans_8x4)(DCTELEM *b, int n);
+ void (*vc1_inv_trans_4x8)(DCTELEM *b, int n);
+ void (*vc1_inv_trans_4x4)(DCTELEM *b, int n);
+ void (*vc1_v_overlap)(uint8_t* src, int stride);
+ void (*vc1_h_overlap)(uint8_t* src, int stride);
+ /* put 8x8 block with bicubic interpolation and quarterpel precision
+ * last argument is actually round value instead of height
+ */
+ op_pixels_func put_vc1_mspel_pixels_tab[16];
} DSPContext;
void dsputil_static_init(void);
void dsputil_init(DSPContext* p, AVCodecContext *avctx);
+int ff_check_alignment(void);
+
/**
* permute block according to permuatation.
* @param last last non zero element in scantable order
#define MM_SSE 0x0008 /* SSE functions */
#define MM_SSE2 0x0010 /* PIV SSE2 functions */
#define MM_3DNOWEXT 0x0020 /* AMD 3DNowExt */
+#define MM_SSE3 0x0040 /* Prescott SSE3 functions */
+#define MM_SSSE3 0x0080 /* Conroe SSSE3 functions */
extern int mm_flags;
void dsputil_init_sh4(DSPContext* c, AVCodecContext *avctx);
-#else
+#elif defined(ARCH_BFIN)
#define DECLARE_ALIGNED_8(t,v) t v __attribute__ ((aligned (8)))
#define STRIDE_ALIGN 8
-#endif
-
-#ifdef __GNUC__
-
-struct unaligned_64 { uint64_t l; } __attribute__((packed));
-struct unaligned_32 { uint32_t l; } __attribute__((packed));
-struct unaligned_16 { uint16_t l; } __attribute__((packed));
-
-#define LD16(a) (((const struct unaligned_16 *) (a))->l)
-#define LD32(a) (((const struct unaligned_32 *) (a))->l)
-#define LD64(a) (((const struct unaligned_64 *) (a))->l)
-
-#define ST16(a, b) (((struct unaligned_16 *) (a))->l) = (b)
-#define ST32(a, b) (((struct unaligned_32 *) (a))->l) = (b)
-
-#else /* __GNUC__ */
+void dsputil_init_bfin(DSPContext* c, AVCodecContext *avctx);
-#define LD16(a) (*((uint16_t*)(a)))
-#define LD32(a) (*((uint32_t*)(a)))
-#define LD64(a) (*((uint64_t*)(a)))
+#else
-#define ST16(a, b) *((uint16_t*)(a)) = (b)
-#define ST32(a, b) *((uint32_t*)(a)) = (b)
+#define DECLARE_ALIGNED_8(t,v) t v __attribute__ ((aligned (8)))
+#define STRIDE_ALIGN 8
-#endif /* !__GNUC__ */
+#endif
/* PSNR */
void get_psnr(uint8_t *orig_image[3], uint8_t *coded_image[3],
FFTSample type */
typedef float FFTSample;
+struct MDCTContext;
+
typedef struct FFTComplex {
FFTSample re, im;
} FFTComplex;
FFTComplex *exptab;
FFTComplex *exptab1; /* only used by SSE code */
void (*fft_calc)(struct FFTContext *s, FFTComplex *z);
+ void (*imdct_calc)(struct MDCTContext *s, FFTSample *output,
+ const FFTSample *input, FFTSample *tmp);
} FFTContext;
int ff_fft_init(FFTContext *s, int nbits, int inverse);
int ff_mdct_init(MDCTContext *s, int nbits, int inverse);
void ff_imdct_calc(MDCTContext *s, FFTSample *output,
const FFTSample *input, FFTSample *tmp);
+void ff_imdct_calc_3dn2(MDCTContext *s, FFTSample *output,
+ const FFTSample *input, FFTSample *tmp);
+void ff_imdct_calc_sse(MDCTContext *s, FFTSample *output,
+ const FFTSample *input, FFTSample *tmp);
void ff_mdct_calc(MDCTContext *s, FFTSample *out,
const FFTSample *input, FFTSample *tmp);
void ff_mdct_end(MDCTContext *s);
return score;\
}
+
+static inline void copy_block2(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
+{
+ int i;
+ for(i=0; i<h; i++)
+ {
+ ST16(dst , LD16(src ));
+ dst+=dstStride;
+ src+=srcStride;
+ }
+}
+
+static inline void copy_block4(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
+{
+ int i;
+ for(i=0; i<h; i++)
+ {
+ ST32(dst , LD32(src ));
+ dst+=dstStride;
+ src+=srcStride;
+ }
+}
+
+static inline void copy_block8(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
+{
+ int i;
+ for(i=0; i<h; i++)
+ {
+ ST32(dst , LD32(src ));
+ ST32(dst+4 , LD32(src+4 ));
+ dst+=dstStride;
+ src+=srcStride;
+ }
+}
+
+static inline void copy_block9(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
+{
+ int i;
+ for(i=0; i<h; i++)
+ {
+ ST32(dst , LD32(src ));
+ ST32(dst+4 , LD32(src+4 ));
+ dst[8]= src[8];
+ dst+=dstStride;
+ src+=srcStride;
+ }
+}
+
+static inline void copy_block16(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
+{
+ int i;
+ for(i=0; i<h; i++)
+ {
+ ST32(dst , LD32(src ));
+ ST32(dst+4 , LD32(src+4 ));
+ ST32(dst+8 , LD32(src+8 ));
+ ST32(dst+12, LD32(src+12));
+ dst+=dstStride;
+ src+=srcStride;
+ }
+}
+
+static inline void copy_block17(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h)
+{
+ int i;
+ for(i=0; i<h; i++)
+ {
+ ST32(dst , LD32(src ));
+ ST32(dst+4 , LD32(src+4 ));
+ ST32(dst+8 , LD32(src+8 ));
+ ST32(dst+12, LD32(src+12));
+ dst[16]= src[16];
+ dst+=dstStride;
+ src+=srcStride;
+ }
+}
+
#endif
projects / ffmpeg / blobdiff