From bb12c2e9c00e3d3c393885849b8fa31cd849dc92 Mon Sep 17 00:00:00 2001 From: Diego Biurrun Date: Thu, 19 Dec 2013 18:28:50 +0100 Subject: [PATCH] dsputil: K&R formatting cosmetics --- libavcodec/dsputil.c | 3423 +++++++++++++++++++++++------------------- 1 file changed, 1881 insertions(+), 1542 deletions(-) diff --git a/libavcodec/dsputil.c b/libavcodec/dsputil.c index ab02cdf4df..8a4594ea8f 100644 --- a/libavcodec/dsputil.c +++ b/libavcodec/dsputil.c @@ -41,7 +41,7 @@ #include "mpegvideo.h" #include "config.h" -uint32_t ff_squareTbl[512] = {0, }; +uint32_t ff_squareTbl[512] = { 0, }; #define BIT_DEPTH 16 #include "dsputil_template.c" @@ -51,11 +51,11 @@ uint32_t ff_squareTbl[512] = {0, }; #include "dsputil_template.c" // 0x7f7f7f7f or 0x7f7f7f7f7f7f7f7f or whatever, depending on the cpu's native arithmetic size -#define pb_7f (~0UL/255 * 0x7f) -#define pb_80 (~0UL/255 * 0x80) +#define pb_7f (~0UL / 255 * 0x7f) +#define pb_80 (~0UL / 255 * 0x80) /* Specific zigzag scan for 248 idct. NOTE that unlike the - specification, we interleave the fields */ + * specification, we interleave the fields */ const uint8_t ff_zigzag248_direct[64] = { 0, 8, 1, 9, 16, 24, 2, 10, 17, 25, 32, 40, 48, 56, 33, 41, @@ -68,7 +68,7 @@ const uint8_t ff_zigzag248_direct[64] = { }; const uint8_t ff_alternate_horizontal_scan[64] = { - 0, 1, 2, 3, 8, 9, 16, 17, + 0, 1, 2, 3, 8, 9, 16, 17, 10, 11, 4, 5, 6, 7, 15, 14, 13, 12, 19, 18, 24, 25, 32, 33, 26, 27, 20, 21, 22, 23, 28, 29, @@ -79,7 +79,7 @@ const uint8_t ff_alternate_horizontal_scan[64] = { }; const uint8_t ff_alternate_vertical_scan[64] = { - 0, 8, 16, 24, 1, 9, 2, 10, + 0, 8, 16, 24, 1, 9, 2, 10, 17, 25, 32, 40, 48, 56, 57, 49, 41, 33, 26, 18, 3, 11, 4, 12, 19, 27, 34, 42, 50, 58, 35, 43, @@ -90,18 +90,18 @@ const uint8_t ff_alternate_vertical_scan[64] = { }; /* Input permutation for the simple_idct_mmx */ -static const uint8_t simple_mmx_permutation[64]={ - 0x00, 0x08, 0x04, 0x09, 0x01, 0x0C, 0x05, 0x0D, - 0x10, 0x18, 0x14, 0x19, 0x11, 0x1C, 0x15, 0x1D, - 0x20, 0x28, 0x24, 0x29, 0x21, 0x2C, 0x25, 0x2D, - 0x12, 0x1A, 0x16, 0x1B, 0x13, 0x1E, 0x17, 0x1F, - 0x02, 0x0A, 0x06, 0x0B, 0x03, 0x0E, 0x07, 0x0F, - 0x30, 0x38, 0x34, 0x39, 0x31, 0x3C, 0x35, 0x3D, - 0x22, 0x2A, 0x26, 0x2B, 0x23, 0x2E, 0x27, 0x2F, - 0x32, 0x3A, 0x36, 0x3B, 0x33, 0x3E, 0x37, 0x3F, +static const uint8_t simple_mmx_permutation[64] = { + 0x00, 0x08, 0x04, 0x09, 0x01, 0x0C, 0x05, 0x0D, + 0x10, 0x18, 0x14, 0x19, 0x11, 0x1C, 0x15, 0x1D, + 0x20, 0x28, 0x24, 0x29, 0x21, 0x2C, 0x25, 0x2D, + 0x12, 0x1A, 0x16, 0x1B, 0x13, 0x1E, 0x17, 0x1F, + 0x02, 0x0A, 0x06, 0x0B, 0x03, 0x0E, 0x07, 0x0F, + 0x30, 0x38, 0x34, 0x39, 0x31, 0x3C, 0x35, 0x3D, + 0x22, 0x2A, 0x26, 0x2B, 0x23, 0x2E, 0x27, 0x2F, + 0x32, 0x3A, 0x36, 0x3B, 0x33, 0x3E, 0x37, 0x3F, }; -static const uint8_t idct_sse2_row_perm[8] = {0, 4, 1, 5, 2, 6, 3, 7}; +static const uint8_t idct_sse2_row_perm[8] = { 0, 4, 1, 5, 2, 6, 3, 7 }; av_cold void ff_init_scantable(uint8_t *permutation, ScanTable *st, const uint8_t *src_scantable) @@ -109,20 +109,21 @@ av_cold void ff_init_scantable(uint8_t *permutation, ScanTable *st, int i; int end; - st->scantable= src_scantable; + st->scantable = src_scantable; - for(i=0; i<64; i++){ + for (i = 0; i < 64; i++) { int j; j = src_scantable[i]; st->permutated[i] = permutation[j]; } - end=-1; - for(i=0; i<64; i++){ + end = -1; + for (i = 0; i < 64; i++) { int j; j = st->permutated[i]; - if(j>end) end=j; - st->raster_end[i]= end; + if (j > end) + end = j; + st->raster_end[i] = end; } } @@ -131,51 +132,52 @@ av_cold void ff_init_scantable_permutation(uint8_t *idct_permutation, { int i; - switch(idct_permutation_type){ + switch (idct_permutation_type) { case FF_NO_IDCT_PERM: - for(i=0; i<64; i++) - idct_permutation[i]= i; + for (i = 0; i < 64; i++) + idct_permutation[i] = i; break; case FF_LIBMPEG2_IDCT_PERM: - for(i=0; i<64; i++) - idct_permutation[i]= (i & 0x38) | ((i & 6) >> 1) | ((i & 1) << 2); + for (i = 0; i < 64; i++) + idct_permutation[i] = (i & 0x38) | ((i & 6) >> 1) | ((i & 1) << 2); break; case FF_SIMPLE_IDCT_PERM: - for(i=0; i<64; i++) - idct_permutation[i]= simple_mmx_permutation[i]; + for (i = 0; i < 64; i++) + idct_permutation[i] = simple_mmx_permutation[i]; break; case FF_TRANSPOSE_IDCT_PERM: - for(i=0; i<64; i++) - idct_permutation[i]= ((i&7)<<3) | (i>>3); + for (i = 0; i < 64; i++) + idct_permutation[i] = ((i & 7) << 3) | (i >> 3); break; case FF_PARTTRANS_IDCT_PERM: - for(i=0; i<64; i++) - idct_permutation[i]= (i&0x24) | ((i&3)<<3) | ((i>>3)&3); + for (i = 0; i < 64; i++) + idct_permutation[i] = (i & 0x24) | ((i & 3) << 3) | ((i >> 3) & 3); break; case FF_SSE2_IDCT_PERM: - for(i=0; i<64; i++) - idct_permutation[i]= (i&0x38) | idct_sse2_row_perm[i&7]; + for (i = 0; i < 64; i++) + idct_permutation[i] = (i & 0x38) | idct_sse2_row_perm[i & 7]; break; default: - av_log(NULL, AV_LOG_ERROR, "Internal error, IDCT permutation not set\n"); + av_log(NULL, AV_LOG_ERROR, + "Internal error, IDCT permutation not set\n"); } } -static int pix_sum_c(uint8_t * pix, int line_size) +static int pix_sum_c(uint8_t *pix, int line_size) { int s, i, j; s = 0; for (i = 0; i < 16; i++) { for (j = 0; j < 16; j += 8) { - s += pix[0]; - s += pix[1]; - s += pix[2]; - s += pix[3]; - s += pix[4]; - s += pix[5]; - s += pix[6]; - s += pix[7]; + s += pix[0]; + s += pix[1]; + s += pix[2]; + s += pix[3]; + s += pix[4]; + s += pix[5]; + s += pix[6]; + s += pix[7]; pix += 8; } pix += line_size - 16; @@ -183,7 +185,7 @@ static int pix_sum_c(uint8_t * pix, int line_size) return s; } -static int pix_norm1_c(uint8_t * pix, int line_size) +static int pix_norm1_c(uint8_t *pix, int line_size) { int s, i, j; uint32_t *sq = ff_squareTbl + 256; @@ -202,26 +204,26 @@ static int pix_norm1_c(uint8_t * pix, int line_size) s += sq[pix[7]]; #else #if HAVE_FAST_64BIT - register uint64_t x=*(uint64_t*)pix; - s += sq[x&0xff]; - s += sq[(x>>8)&0xff]; - s += sq[(x>>16)&0xff]; - s += sq[(x>>24)&0xff]; - s += sq[(x>>32)&0xff]; - s += sq[(x>>40)&0xff]; - s += sq[(x>>48)&0xff]; - s += sq[(x>>56)&0xff]; + register uint64_t x = *(uint64_t *) pix; + s += sq[x & 0xff]; + s += sq[(x >> 8) & 0xff]; + s += sq[(x >> 16) & 0xff]; + s += sq[(x >> 24) & 0xff]; + s += sq[(x >> 32) & 0xff]; + s += sq[(x >> 40) & 0xff]; + s += sq[(x >> 48) & 0xff]; + s += sq[(x >> 56) & 0xff]; #else - register uint32_t x=*(uint32_t*)pix; - s += sq[x&0xff]; - s += sq[(x>>8)&0xff]; - s += sq[(x>>16)&0xff]; - s += sq[(x>>24)&0xff]; - x=*(uint32_t*)(pix+4); - s += sq[x&0xff]; - s += sq[(x>>8)&0xff]; - s += sq[(x>>16)&0xff]; - s += sq[(x>>24)&0xff]; + register uint32_t x = *(uint32_t *) pix; + s += sq[x & 0xff]; + s += sq[(x >> 8) & 0xff]; + s += sq[(x >> 16) & 0xff]; + s += sq[(x >> 24) & 0xff]; + x = *(uint32_t *) (pix + 4); + s += sq[x & 0xff]; + s += sq[(x >> 8) & 0xff]; + s += sq[(x >> 16) & 0xff]; + s += sq[(x >> 24) & 0xff]; #endif #endif pix += 8; @@ -231,22 +233,22 @@ static int pix_norm1_c(uint8_t * pix, int line_size) return s; } -static void bswap_buf(uint32_t *dst, const uint32_t *src, int w){ +static void bswap_buf(uint32_t *dst, const uint32_t *src, int w) +{ int i; - for(i=0; i+8<=w; i+=8){ - dst[i+0]= av_bswap32(src[i+0]); - dst[i+1]= av_bswap32(src[i+1]); - dst[i+2]= av_bswap32(src[i+2]); - dst[i+3]= av_bswap32(src[i+3]); - dst[i+4]= av_bswap32(src[i+4]); - dst[i+5]= av_bswap32(src[i+5]); - dst[i+6]= av_bswap32(src[i+6]); - dst[i+7]= av_bswap32(src[i+7]); - } - for(;i 127) *pixels = 255; else - *pixels = (uint8_t)(*block + 128); + *pixels = (uint8_t) (*block + 128); block++; pixels++; } @@ -386,13 +388,12 @@ static void put_signed_pixels_clamped_c(const int16_t *block, } } -static void add_pixels8_c(uint8_t *restrict pixels, - int16_t *block, +static void add_pixels8_c(uint8_t *restrict pixels, int16_t *block, int line_size) { int i; - for(i=0;i<8;i++) { + for (i = 0; i < 8; i++) { pixels[0] += block[0]; pixels[1] += block[1]; pixels[2] += block[2]; @@ -401,8 +402,8 @@ static void add_pixels8_c(uint8_t *restrict pixels, pixels[5] += block[5]; pixels[6] += block[6]; pixels[7] += block[7]; - pixels += line_size; - block += 8; + pixels += line_size; + block += 8; } } @@ -412,7 +413,7 @@ static void add_pixels_clamped_c(const int16_t *block, uint8_t *restrict pixels, int i; /* read the pixels */ - for(i=0;i<8;i++) { + for (i = 0; i < 8; i++) { pixels[0] = av_clip_uint8(pixels[0] + block[0]); pixels[1] = av_clip_uint8(pixels[1] + block[1]); pixels[2] = av_clip_uint8(pixels[2] + block[2]); @@ -421,16 +422,17 @@ static void add_pixels_clamped_c(const int16_t *block, uint8_t *restrict pixels, pixels[5] = av_clip_uint8(pixels[5] + block[5]); pixels[6] = av_clip_uint8(pixels[6] + block[6]); pixels[7] = av_clip_uint8(pixels[7] + block[7]); - pixels += line_size; - block += 8; + pixels += line_size; + block += 8; } } static int sum_abs_dctelem_c(int16_t *block) { - int sum=0, i; - for(i=0; i<64; i++) - sum+= FFABS(block[i]); + int sum = 0, i; + + for (i = 0; i < 64; i++) + sum += FFABS(block[i]); return sum; } @@ -454,810 +456,1028 @@ static void fill_block8_c(uint8_t *block, uint8_t value, int line_size, int h) } } -#define avg2(a,b) ((a+b+1)>>1) -#define avg4(a,b,c,d) ((a+b+c+d+2)>>2) +#define avg2(a, b) ((a + b + 1) >> 1) +#define avg4(a, b, c, d) ((a + b + c + d + 2) >> 2) -static void gmc1_c(uint8_t *dst, uint8_t *src, int stride, int h, int x16, int y16, int rounder) +static void gmc1_c(uint8_t *dst, uint8_t *src, int stride, int h, + int x16, int y16, int rounder) { - const int A=(16-x16)*(16-y16); - const int B=( x16)*(16-y16); - const int C=(16-x16)*( y16); - const int D=( x16)*( y16); + const int A = (16 - x16) * (16 - y16); + const int B = (x16) * (16 - y16); + const int C = (16 - x16) * (y16); + const int D = (x16) * (y16); int i; - for(i=0; i>8; - dst[1]= (A*src[1] + B*src[2] + C*src[stride+1] + D*src[stride+2] + rounder)>>8; - dst[2]= (A*src[2] + B*src[3] + C*src[stride+2] + D*src[stride+3] + rounder)>>8; - dst[3]= (A*src[3] + B*src[4] + C*src[stride+3] + D*src[stride+4] + rounder)>>8; - dst[4]= (A*src[4] + B*src[5] + C*src[stride+4] + D*src[stride+5] + rounder)>>8; - dst[5]= (A*src[5] + B*src[6] + C*src[stride+5] + D*src[stride+6] + rounder)>>8; - dst[6]= (A*src[6] + B*src[7] + C*src[stride+6] + D*src[stride+7] + rounder)>>8; - dst[7]= (A*src[7] + B*src[8] + C*src[stride+7] + D*src[stride+8] + rounder)>>8; - dst+= stride; - src+= stride; + for (i = 0; i < h; i++) { + dst[0] = (A * src[0] + B * src[1] + C * src[stride + 0] + D * src[stride + 1] + rounder) >> 8; + dst[1] = (A * src[1] + B * src[2] + C * src[stride + 1] + D * src[stride + 2] + rounder) >> 8; + dst[2] = (A * src[2] + B * src[3] + C * src[stride + 2] + D * src[stride + 3] + rounder) >> 8; + dst[3] = (A * src[3] + B * src[4] + C * src[stride + 3] + D * src[stride + 4] + rounder) >> 8; + dst[4] = (A * src[4] + B * src[5] + C * src[stride + 4] + D * src[stride + 5] + rounder) >> 8; + dst[5] = (A * src[5] + B * src[6] + C * src[stride + 5] + D * src[stride + 6] + rounder) >> 8; + dst[6] = (A * src[6] + B * src[7] + C * src[stride + 6] + D * src[stride + 7] + rounder) >> 8; + dst[7] = (A * src[7] + B * src[8] + C * src[stride + 7] + D * src[stride + 8] + rounder) >> 8; + dst += stride; + src += stride; } } 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) + int dxx, int dxy, int dyx, int dyy, int shift, int r, + int width, int height) { int y, vx, vy; - const int s= 1<>16; - src_y= vy>>16; - frac_x= src_x&(s-1); - frac_y= src_y&(s-1); - src_x>>=shift; - src_y>>=shift; + src_x = vx >> 16; + src_y = vy >> 16; + frac_x = src_x & (s - 1); + frac_y = src_y & (s - 1); + src_x >>= shift; + src_y >>= shift; - if((unsigned)src_x < width){ - if((unsigned)src_y < height){ - index= src_x + src_y*stride; - dst[y*stride + x]= ( ( src[index ]*(s-frac_x) - + src[index +1]* frac_x )*(s-frac_y) - + ( src[index+stride ]*(s-frac_x) - + src[index+stride+1]* frac_x )* frac_y - + r)>>(shift*2); - }else{ - index= src_x + av_clip(src_y, 0, height)*stride; - dst[y*stride + x]= ( ( src[index ]*(s-frac_x) - + src[index +1]* frac_x )*s - + r)>>(shift*2); + if ((unsigned) src_x < width) { + if ((unsigned) src_y < height) { + index = src_x + src_y * stride; + dst[y * stride + x] = + ((src[index] * (s - frac_x) + + src[index + 1] * frac_x) * (s - frac_y) + + (src[index + stride] * (s - frac_x) + + src[index + stride + 1] * frac_x) * frac_y + + r) >> (shift * 2); + } else { + index = src_x + av_clip(src_y, 0, height) * stride; + dst[y * stride + x] = + ((src[index] * (s - frac_x) + + src[index + 1] * frac_x) * s + + r) >> (shift * 2); } - }else{ - if((unsigned)src_y < height){ - index= av_clip(src_x, 0, width) + src_y*stride; - dst[y*stride + x]= ( ( src[index ]*(s-frac_y) - + src[index+stride ]* frac_y )*s - + r)>>(shift*2); - }else{ - index= av_clip(src_x, 0, width) + av_clip(src_y, 0, height)*stride; - dst[y*stride + x]= src[index ]; + } else { + if ((unsigned) src_y < height) { + index = av_clip(src_x, 0, width) + src_y * stride; + dst[y * stride + x] = + ((src[index] * (s - frac_y) + + src[index + stride] * frac_y) * s + + r) >> (shift * 2); + } else { + index = av_clip(src_x, 0, width) + + av_clip(src_y, 0, height) * stride; + dst[y * stride + x] = src[index]; } } - vx+= dxx; - vy+= dyx; + vx += dxx; + vy += dyx; } ox += dxy; oy += dyy; } } -static inline void put_tpel_pixels_mc00_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - switch(width){ - case 2: put_pixels2_8_c (dst, src, stride, height); break; - case 4: put_pixels4_8_c (dst, src, stride, height); break; - case 8: put_pixels8_8_c (dst, src, stride, height); break; - case 16:put_pixels16_8_c(dst, src, stride, height); break; +static inline void put_tpel_pixels_mc00_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + switch (width) { + case 2: + put_pixels2_8_c(dst, src, stride, height); + break; + case 4: + put_pixels4_8_c(dst, src, stride, height); + break; + case 8: + put_pixels8_8_c(dst, src, stride, height); + break; + case 16: + put_pixels16_8_c(dst, src, stride, height); + break; } } -static inline void put_tpel_pixels_mc10_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (683*(2*src[j] + src[j+1] + 1)) >> 11; - } - src += stride; - dst += stride; +static inline void put_tpel_pixels_mc10_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = ((2 * src[j] + src[j + 1] + 1) * + 683) >> 11; + src += stride; + dst += stride; } } -static inline void put_tpel_pixels_mc20_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (683*(src[j] + 2*src[j+1] + 1)) >> 11; - } - src += stride; - dst += stride; +static inline void put_tpel_pixels_mc20_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = ((src[j] + 2 * src[j + 1] + 1) * + 683) >> 11; + src += stride; + dst += stride; } } -static inline void put_tpel_pixels_mc01_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (683*(2*src[j] + src[j+stride] + 1)) >> 11; - } - src += stride; - dst += stride; +static inline void put_tpel_pixels_mc01_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = ((2 * src[j] + src[j + stride] + 1) * + 683) >> 11; + src += stride; + dst += stride; } } -static inline void put_tpel_pixels_mc11_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (2731*(4*src[j] + 3*src[j+1] + 3*src[j+stride] + 2*src[j+stride+1] + 6)) >> 15; - } - src += stride; - dst += stride; +static inline void put_tpel_pixels_mc11_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = ((4 * src[j] + 3 * src[j + 1] + + 3 * src[j + stride] + 2 * src[j + stride + 1] + 6) * + 2731) >> 15; + src += stride; + dst += stride; } } -static inline void put_tpel_pixels_mc12_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (2731*(3*src[j] + 2*src[j+1] + 4*src[j+stride] + 3*src[j+stride+1] + 6)) >> 15; - } - src += stride; - dst += stride; +static inline void put_tpel_pixels_mc12_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = ((3 * src[j] + 2 * src[j + 1] + + 4 * src[j + stride] + 3 * src[j + stride + 1] + 6) * + 2731) >> 15; + src += stride; + dst += stride; } } -static inline void put_tpel_pixels_mc02_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (683*(src[j] + 2*src[j+stride] + 1)) >> 11; - } - src += stride; - dst += stride; +static inline void put_tpel_pixels_mc02_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = ((src[j] + 2 * src[j + stride] + 1) * + 683) >> 11; + src += stride; + dst += stride; } } -static inline void put_tpel_pixels_mc21_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (2731*(3*src[j] + 4*src[j+1] + 2*src[j+stride] + 3*src[j+stride+1] + 6)) >> 15; - } - src += stride; - dst += stride; +static inline void put_tpel_pixels_mc21_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = ((3 * src[j] + 4 * src[j + 1] + + 2 * src[j + stride] + 3 * src[j + stride + 1] + 6) * + 2731) >> 15; + src += stride; + dst += stride; } } -static inline void put_tpel_pixels_mc22_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (2731*(2*src[j] + 3*src[j+1] + 3*src[j+stride] + 4*src[j+stride+1] + 6)) >> 15; - } - src += stride; - dst += stride; +static inline void put_tpel_pixels_mc22_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = ((2 * src[j] + 3 * src[j + 1] + + 3 * src[j + stride] + 4 * src[j + stride + 1] + 6) * + 2731) >> 15; + src += stride; + dst += stride; } } -static inline void avg_tpel_pixels_mc00_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - switch(width){ - case 2: avg_pixels2_8_c (dst, src, stride, height); break; - case 4: avg_pixels4_8_c (dst, src, stride, height); break; - case 8: avg_pixels8_8_c (dst, src, stride, height); break; - case 16:avg_pixels16_8_c(dst, src, stride, height); break; +static inline void avg_tpel_pixels_mc00_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + switch (width) { + case 2: + avg_pixels2_8_c(dst, src, stride, height); + break; + case 4: + avg_pixels4_8_c(dst, src, stride, height); + break; + case 8: + avg_pixels8_8_c(dst, src, stride, height); + break; + case 16: + avg_pixels16_8_c(dst, src, stride, height); + break; } } -static inline void avg_tpel_pixels_mc10_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (dst[j] + ((683*(2*src[j] + src[j+1] + 1)) >> 11) + 1) >> 1; - } - src += stride; - dst += stride; +static inline void avg_tpel_pixels_mc10_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = (dst[j] + + (((2 * src[j] + src[j + 1] + 1) * + 683) >> 11) + 1) >> 1; + src += stride; + dst += stride; } } -static inline void avg_tpel_pixels_mc20_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (dst[j] + ((683*(src[j] + 2*src[j+1] + 1)) >> 11) + 1) >> 1; - } - src += stride; - dst += stride; +static inline void avg_tpel_pixels_mc20_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = (dst[j] + + (((src[j] + 2 * src[j + 1] + 1) * + 683) >> 11) + 1) >> 1; + src += stride; + dst += stride; } } -static inline void avg_tpel_pixels_mc01_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (dst[j] + ((683*(2*src[j] + src[j+stride] + 1)) >> 11) + 1) >> 1; - } - src += stride; - dst += stride; +static inline void avg_tpel_pixels_mc01_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = (dst[j] + + (((2 * src[j] + src[j + stride] + 1) * + 683) >> 11) + 1) >> 1; + src += stride; + dst += stride; } } -static inline void avg_tpel_pixels_mc11_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (dst[j] + ((2731*(4*src[j] + 3*src[j+1] + 3*src[j+stride] + 2*src[j+stride+1] + 6)) >> 15) + 1) >> 1; - } - src += stride; - dst += stride; +static inline void avg_tpel_pixels_mc11_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = (dst[j] + + (((4 * src[j] + 3 * src[j + 1] + + 3 * src[j + stride] + 2 * src[j + stride + 1] + 6) * + 2731) >> 15) + 1) >> 1; + src += stride; + dst += stride; } } -static inline void avg_tpel_pixels_mc12_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (dst[j] + ((2731*(3*src[j] + 2*src[j+1] + 4*src[j+stride] + 3*src[j+stride+1] + 6)) >> 15) + 1) >> 1; - } - src += stride; - dst += stride; +static inline void avg_tpel_pixels_mc12_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = (dst[j] + + (((3 * src[j] + 2 * src[j + 1] + + 4 * src[j + stride] + 3 * src[j + stride + 1] + 6) * + 2731) >> 15) + 1) >> 1; + src += stride; + dst += stride; } } -static inline void avg_tpel_pixels_mc02_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (dst[j] + ((683*(src[j] + 2*src[j+stride] + 1)) >> 11) + 1) >> 1; - } - src += stride; - dst += stride; +static inline void avg_tpel_pixels_mc02_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = (dst[j] + + (((src[j] + 2 * src[j + stride] + 1) * + 683) >> 11) + 1) >> 1; + src += stride; + dst += stride; } } -static inline void avg_tpel_pixels_mc21_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (dst[j] + ((2731*(3*src[j] + 4*src[j+1] + 2*src[j+stride] + 3*src[j+stride+1] + 6)) >> 15) + 1) >> 1; - } - src += stride; - dst += stride; +static inline void avg_tpel_pixels_mc21_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = (dst[j] + + (((3 * src[j] + 4 * src[j + 1] + + 2 * src[j + stride] + 3 * src[j + stride + 1] + 6) * + 2731) >> 15) + 1) >> 1; + src += stride; + dst += stride; } } -static inline void avg_tpel_pixels_mc22_c(uint8_t *dst, const uint8_t *src, int stride, int width, int height){ - int i,j; - for (i=0; i < height; i++) { - for (j=0; j < width; j++) { - dst[j] = (dst[j] + ((2731*(2*src[j] + 3*src[j+1] + 3*src[j+stride] + 4*src[j+stride+1] + 6)) >> 15) + 1) >> 1; - } - src += stride; - dst += stride; +static inline void avg_tpel_pixels_mc22_c(uint8_t *dst, const uint8_t *src, + int stride, int width, int height) +{ + int i, j; + + for (i = 0; i < height; i++) { + for (j = 0; j < width; j++) + dst[j] = (dst[j] + + (((2 * src[j] + 3 * src[j + 1] + + 3 * src[j + stride] + 4 * src[j + stride + 1] + 6) * + 2731) >> 15) + 1) >> 1; + src += stride; + dst += stride; } } -#define QPEL_MC(r, OPNAME, RND, OP) \ -static void OPNAME ## mpeg4_qpel8_h_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h){\ - const uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\ - int i;\ - for(i=0; i>5]+1)>>1) -#define op_avg_no_rnd(a, b) a = (((a)+cm[((b) + 15)>>5])>>1) -#define op_put(a, b) a = cm[((b) + 16)>>5] -#define op_put_no_rnd(a, b) a = cm[((b) + 15)>>5] +#define op_avg(a, b) a = (((a) + cm[((b) + 16) >> 5] + 1) >> 1) +#define op_avg_no_rnd(a, b) a = (((a) + cm[((b) + 15) >> 5]) >> 1) +#define op_put(a, b) a = cm[((b) + 16) >> 5] +#define op_put_no_rnd(a, b) a = cm[((b) + 15) >> 5] -QPEL_MC(0, put_ , _ , op_put) +QPEL_MC(0, put_, _, op_put) QPEL_MC(1, put_no_rnd_, _no_rnd_, op_put_no_rnd) -QPEL_MC(0, avg_ , _ , op_avg) +QPEL_MC(0, avg_, _, op_avg) #undef op_avg #undef op_put @@ -1267,41 +1487,46 @@ void ff_put_pixels8x8_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride) { put_pixels8_8_c(dst, src, stride, 8); } + void ff_avg_pixels8x8_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride) { avg_pixels8_8_c(dst, src, stride, 8); } + void ff_put_pixels16x16_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride) { put_pixels16_8_c(dst, src, stride, 16); } + void ff_avg_pixels16x16_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride) { avg_pixels16_8_c(dst, src, stride, 16); } -#define put_qpel8_mc00_c ff_put_pixels8x8_c -#define avg_qpel8_mc00_c ff_avg_pixels8x8_c -#define put_qpel16_mc00_c ff_put_pixels16x16_c -#define avg_qpel16_mc00_c ff_avg_pixels16x16_c +#define put_qpel8_mc00_c ff_put_pixels8x8_c +#define avg_qpel8_mc00_c ff_avg_pixels8x8_c +#define put_qpel16_mc00_c ff_put_pixels16x16_c +#define avg_qpel16_mc00_c ff_avg_pixels16x16_c #define put_no_rnd_qpel8_mc00_c ff_put_pixels8x8_c #define put_no_rnd_qpel16_mc00_c ff_put_pixels16x16_c -static void wmv2_mspel8_h_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int h){ +static void wmv2_mspel8_h_lowpass(uint8_t *dst, uint8_t *src, + int dstStride, int srcStride, int h) +{ const uint8_t *cm = ff_cropTbl + MAX_NEG_CROP; int i; - for(i=0; i>4]; - dst[1]= cm[(9*(src[1] + src[2]) - (src[ 0] + src[3]) + 8)>>4]; - dst[2]= cm[(9*(src[2] + src[3]) - (src[ 1] + src[4]) + 8)>>4]; - dst[3]= cm[(9*(src[3] + src[4]) - (src[ 2] + src[5]) + 8)>>4]; - dst[4]= cm[(9*(src[4] + src[5]) - (src[ 3] + src[6]) + 8)>>4]; - dst[5]= cm[(9*(src[5] + src[6]) - (src[ 4] + src[7]) + 8)>>4]; - dst[6]= cm[(9*(src[6] + src[7]) - (src[ 5] + src[8]) + 8)>>4]; - dst[7]= cm[(9*(src[7] + src[8]) - (src[ 6] + src[9]) + 8)>>4]; - dst+=dstStride; - src+=srcStride; + for (i = 0; i < h; i++) { + dst[0] = cm[(9 * (src[0] + src[1]) - (src[-1] + src[2]) + 8) >> 4]; + dst[1] = cm[(9 * (src[1] + src[2]) - (src[0] + src[3]) + 8) >> 4]; + dst[2] = cm[(9 * (src[2] + src[3]) - (src[1] + src[4]) + 8) >> 4]; + dst[3] = cm[(9 * (src[3] + src[4]) - (src[2] + src[5]) + 8) >> 4]; + dst[4] = cm[(9 * (src[4] + src[5]) - (src[3] + src[6]) + 8) >> 4]; + dst[5] = cm[(9 * (src[5] + src[6]) - (src[4] + src[7]) + 8) >> 4]; + dst[6] = cm[(9 * (src[6] + src[7]) - (src[5] + src[8]) + 8) >> 4]; + dst[7] = cm[(9 * (src[7] + src[8]) - (src[6] + src[9]) + 8) >> 4]; + dst += dstStride; + src += srcStride; } } @@ -1310,44 +1535,49 @@ void ff_put_rv40_qpel16_mc33_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride) { put_pixels16_xy2_8_c(dst, src, stride, 16); } + void ff_avg_rv40_qpel16_mc33_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride) { avg_pixels16_xy2_8_c(dst, src, stride, 16); } + void ff_put_rv40_qpel8_mc33_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride) { put_pixels8_xy2_8_c(dst, src, stride, 8); } + void ff_avg_rv40_qpel8_mc33_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride) { avg_pixels8_xy2_8_c(dst, src, stride, 8); } #endif /* CONFIG_RV40_DECODER */ -static void wmv2_mspel8_v_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int srcStride, int w){ +static void wmv2_mspel8_v_lowpass(uint8_t *dst, uint8_t *src, + int dstStride, int srcStride, int w) +{ const uint8_t *cm = ff_cropTbl + MAX_NEG_CROP; int i; - for(i=0; i>4]; - dst[1*dstStride]= cm[(9*(src1 + src2) - (src0 + src3) + 8)>>4]; - dst[2*dstStride]= cm[(9*(src2 + src3) - (src1 + src4) + 8)>>4]; - dst[3*dstStride]= cm[(9*(src3 + src4) - (src2 + src5) + 8)>>4]; - dst[4*dstStride]= cm[(9*(src4 + src5) - (src3 + src6) + 8)>>4]; - dst[5*dstStride]= cm[(9*(src5 + src6) - (src4 + src7) + 8)>>4]; - dst[6*dstStride]= cm[(9*(src6 + src7) - (src5 + src8) + 8)>>4]; - dst[7*dstStride]= cm[(9*(src7 + src8) - (src6 + src9) + 8)>>4]; + for (i = 0; i < w; i++) { + const int src_1 = src[-srcStride]; + const int src0 = src[0]; + const int src1 = src[srcStride]; + const int src2 = src[2 * srcStride]; + const int src3 = src[3 * srcStride]; + const int src4 = src[4 * srcStride]; + const int src5 = src[5 * srcStride]; + const int src6 = src[6 * srcStride]; + const int src7 = src[7 * srcStride]; + const int src8 = src[8 * srcStride]; + const int src9 = src[9 * srcStride]; + dst[0 * dstStride] = cm[(9 * (src0 + src1) - (src_1 + src2) + 8) >> 4]; + dst[1 * dstStride] = cm[(9 * (src1 + src2) - (src0 + src3) + 8) >> 4]; + dst[2 * dstStride] = cm[(9 * (src2 + src3) - (src1 + src4) + 8) >> 4]; + dst[3 * dstStride] = cm[(9 * (src3 + src4) - (src2 + src5) + 8) >> 4]; + dst[4 * dstStride] = cm[(9 * (src4 + src5) - (src3 + src6) + 8) >> 4]; + dst[5 * dstStride] = cm[(9 * (src5 + src6) - (src4 + src7) + 8) >> 4]; + dst[6 * dstStride] = cm[(9 * (src6 + src7) - (src5 + src8) + 8) >> 4]; + dst[7 * dstStride] = cm[(9 * (src7 + src8) - (src6 + src9) + 8) >> 4]; src++; dst++; } @@ -1356,6 +1586,7 @@ static void wmv2_mspel8_v_lowpass(uint8_t *dst, uint8_t *src, int dstStride, int static void put_mspel8_mc10_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride) { uint8_t half[64]; + wmv2_mspel8_h_lowpass(half, src, 8, stride, 8); put_pixels8_l2_8(dst, src, half, stride, stride, 8, 8); } @@ -1368,8 +1599,9 @@ static void put_mspel8_mc20_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride) static void put_mspel8_mc30_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride) { uint8_t half[64]; + wmv2_mspel8_h_lowpass(half, src, 8, stride, 8); - put_pixels8_l2_8(dst, src+1, half, stride, stride, 8, 8); + put_pixels8_l2_8(dst, src + 1, half, stride, stride, 8, 8); } static void put_mspel8_mc02_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride) @@ -1382,107 +1614,115 @@ static void put_mspel8_mc12_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride) uint8_t halfH[88]; uint8_t halfV[64]; uint8_t halfHV[64]; - wmv2_mspel8_h_lowpass(halfH, src-stride, 8, stride, 11); + + wmv2_mspel8_h_lowpass(halfH, src - stride, 8, stride, 11); wmv2_mspel8_v_lowpass(halfV, src, 8, stride, 8); - wmv2_mspel8_v_lowpass(halfHV, halfH+8, 8, 8, 8); + wmv2_mspel8_v_lowpass(halfHV, halfH + 8, 8, 8, 8); put_pixels8_l2_8(dst, halfV, halfHV, stride, 8, 8, 8); } + static void put_mspel8_mc32_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride) { uint8_t halfH[88]; uint8_t halfV[64]; uint8_t halfHV[64]; - wmv2_mspel8_h_lowpass(halfH, src-stride, 8, stride, 11); - wmv2_mspel8_v_lowpass(halfV, src+1, 8, stride, 8); - wmv2_mspel8_v_lowpass(halfHV, halfH+8, 8, 8, 8); + + wmv2_mspel8_h_lowpass(halfH, src - stride, 8, stride, 11); + wmv2_mspel8_v_lowpass(halfV, src + 1, 8, stride, 8); + wmv2_mspel8_v_lowpass(halfHV, halfH + 8, 8, 8, 8); put_pixels8_l2_8(dst, halfV, halfHV, stride, 8, 8, 8); } + static void put_mspel8_mc22_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride) { uint8_t halfH[88]; - wmv2_mspel8_h_lowpass(halfH, src-stride, 8, stride, 11); - wmv2_mspel8_v_lowpass(dst, halfH+8, stride, 8, 8); + + wmv2_mspel8_h_lowpass(halfH, src - stride, 8, stride, 11); + wmv2_mspel8_v_lowpass(dst, halfH + 8, stride, 8, 8); } -static inline int pix_abs16_c(void *v, uint8_t *pix1, uint8_t *pix2, int line_size, int h) +static inline int pix_abs16_c(void *v, uint8_t *pix1, uint8_t *pix2, + int line_size, int h) { int s, i; s = 0; - for(i=0;iavctx->nsse_weight; - else return score1 + FFABS(score2)*8; + if (c) + return score1 + FFABS(score2) * c->avctx->nsse_weight; + else + return score1 + FFABS(score2) * 8; } -static int nsse8_c(void *v, uint8_t *s1, uint8_t *s2, int stride, int h){ +static int nsse8_c(void *v, uint8_t *s1, uint8_t *s2, int stride, int h) +{ MpegEncContext *c = v; - int score1=0; - int score2=0; - int x,y; + int score1 = 0; + int score2 = 0; + int x, y; - for(y=0; yavctx->nsse_weight; - else return score1 + FFABS(score2)*8; + if (c) + return score1 + FFABS(score2) * c->avctx->nsse_weight; + else + return score1 + FFABS(score2) * 8; } -static int try_8x8basis_c(int16_t rem[64], int16_t weight[64], int16_t basis[64], int scale){ +static int try_8x8basis_c(int16_t rem[64], int16_t weight[64], + int16_t basis[64], int scale) +{ int i; - unsigned int sum=0; + unsigned int sum = 0; - for(i=0; i<8*8; i++){ - int b= rem[i] + ((basis[i]*scale + (1<<(BASIS_SHIFT - RECON_SHIFT-1)))>>(BASIS_SHIFT - RECON_SHIFT)); - int w= weight[i]; - b>>= RECON_SHIFT; - assert(-512> + (BASIS_SHIFT - RECON_SHIFT)); + int w = weight[i]; + b >>= RECON_SHIFT; + assert(-512 < b && b < 512); - sum += (w*b)*(w*b)>>4; + sum += (w * b) * (w * b) >> 4; } - return sum>>2; + return sum >> 2; } -static void add_8x8basis_c(int16_t rem[64], int16_t basis[64], int scale){ +static void add_8x8basis_c(int16_t rem[64], int16_t basis[64], int scale) +{ int i; - for(i=0; i<8*8; i++){ - rem[i] += (basis[i]*scale + (1<<(BASIS_SHIFT - RECON_SHIFT-1)))>>(BASIS_SHIFT - RECON_SHIFT); - } + for (i = 0; i < 8 * 8; i++) + rem[i] += (basis[i] * scale + + (1 << (BASIS_SHIFT - RECON_SHIFT - 1))) >> + (BASIS_SHIFT - RECON_SHIFT); } -static int zero_cmp(void *s, uint8_t *a, uint8_t *b, int stride, int h){ +static int zero_cmp(void *s, uint8_t *a, uint8_t *b, int stride, int h) +{ return 0; } -void ff_set_cmp(DSPContext* c, me_cmp_func *cmp, int type){ +void ff_set_cmp(DSPContext *c, me_cmp_func *cmp, int type) +{ int i; - memset(cmp, 0, sizeof(void*)*6); + memset(cmp, 0, sizeof(void *) * 6); - for(i=0; i<6; i++){ - switch(type&0xFF){ + for (i = 0; i < 6; i++) { + switch (type & 0xFF) { case FF_CMP_SAD: - cmp[i]= c->sad[i]; + cmp[i] = c->sad[i]; break; case FF_CMP_SATD: - cmp[i]= c->hadamard8_diff[i]; + cmp[i] = c->hadamard8_diff[i]; break; case FF_CMP_SSE: - cmp[i]= c->sse[i]; + cmp[i] = c->sse[i]; break; case FF_CMP_DCT: - cmp[i]= c->dct_sad[i]; + cmp[i] = c->dct_sad[i]; break; case FF_CMP_DCT264: - cmp[i]= c->dct264_sad[i]; + cmp[i] = c->dct264_sad[i]; break; case FF_CMP_DCTMAX: - cmp[i]= c->dct_max[i]; + cmp[i] = c->dct_max[i]; break; case FF_CMP_PSNR: - cmp[i]= c->quant_psnr[i]; + cmp[i] = c->quant_psnr[i]; break; case FF_CMP_BIT: - cmp[i]= c->bit[i]; + cmp[i] = c->bit[i]; break; case FF_CMP_RD: - cmp[i]= c->rd[i]; + cmp[i] = c->rd[i]; break; case FF_CMP_VSAD: - cmp[i]= c->vsad[i]; + cmp[i] = c->vsad[i]; break; case FF_CMP_VSSE: - cmp[i]= c->vsse[i]; + cmp[i] = c->vsse[i]; break; case FF_CMP_ZERO: - cmp[i]= zero_cmp; + cmp[i] = zero_cmp; break; case FF_CMP_NSSE: - cmp[i]= c->nsse[i]; + cmp[i] = c->nsse[i]; break; default: - av_log(NULL, AV_LOG_ERROR,"internal error in cmp function selection\n"); + av_log(NULL, AV_LOG_ERROR, + "internal error in cmp function selection\n"); } } } -static void add_bytes_c(uint8_t *dst, uint8_t *src, int w){ +static void add_bytes_c(uint8_t *dst, uint8_t *src, int w) +{ long i; + for (i = 0; i <= w - (int) sizeof(long); i += sizeof(long)) { - long a = *(long*)(src+i); - long b = *(long*)(dst+i); - *(long*)(dst+i) = ((a&pb_7f) + (b&pb_7f)) ^ ((a^b)&pb_80); + long a = *(long *) (src + i); + long b = *(long *) (dst + i); + *(long *) (dst + i) = ((a & pb_7f) + (b & pb_7f)) ^ ((a ^ b) & pb_80); } - for(; idsp.diff_pixels(temp, src1, src2, stride); s->dsp.fdct(temp); @@ -1986,161 +2275,172 @@ static int dct_sad8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *src2 } #if CONFIG_GPL -#define DCT8_1D {\ - const int s07 = SRC(0) + SRC(7);\ - const int s16 = SRC(1) + SRC(6);\ - const int s25 = SRC(2) + SRC(5);\ - const int s34 = SRC(3) + SRC(4);\ - const int a0 = s07 + s34;\ - const int a1 = s16 + s25;\ - const int a2 = s07 - s34;\ - const int a3 = s16 - s25;\ - const int d07 = SRC(0) - SRC(7);\ - const int d16 = SRC(1) - SRC(6);\ - const int d25 = SRC(2) - SRC(5);\ - const int d34 = SRC(3) - SRC(4);\ - const int a4 = d16 + d25 + (d07 + (d07>>1));\ - const int a5 = d07 - d34 - (d25 + (d25>>1));\ - const int a6 = d07 + d34 - (d16 + (d16>>1));\ - const int a7 = d16 - d25 + (d34 + (d34>>1));\ - DST(0, a0 + a1 ) ;\ - DST(1, a4 + (a7>>2)) ;\ - DST(2, a2 + (a3>>1)) ;\ - DST(3, a5 + (a6>>2)) ;\ - DST(4, a0 - a1 ) ;\ - DST(5, a6 - (a5>>2)) ;\ - DST(6, (a2>>1) - a3 ) ;\ - DST(7, (a4>>2) - a7 ) ;\ -} +#define DCT8_1D \ + { \ + const int s07 = SRC(0) + SRC(7); \ + const int s16 = SRC(1) + SRC(6); \ + const int s25 = SRC(2) + SRC(5); \ + const int s34 = SRC(3) + SRC(4); \ + const int a0 = s07 + s34; \ + const int a1 = s16 + s25; \ + const int a2 = s07 - s34; \ + const int a3 = s16 - s25; \ + const int d07 = SRC(0) - SRC(7); \ + const int d16 = SRC(1) - SRC(6); \ + const int d25 = SRC(2) - SRC(5); \ + const int d34 = SRC(3) - SRC(4); \ + const int a4 = d16 + d25 + (d07 + (d07 >> 1)); \ + const int a5 = d07 - d34 - (d25 + (d25 >> 1)); \ + const int a6 = d07 + d34 - (d16 + (d16 >> 1)); \ + const int a7 = d16 - d25 + (d34 + (d34 >> 1)); \ + DST(0, a0 + a1); \ + DST(1, a4 + (a7 >> 2)); \ + DST(2, a2 + (a3 >> 1)); \ + DST(3, a5 + (a6 >> 2)); \ + DST(4, a0 - a1); \ + DST(5, a6 - (a5 >> 2)); \ + DST(6, (a2 >> 1) - a3); \ + DST(7, (a4 >> 2) - a7); \ + } -static int dct264_sad8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *src2, int stride, int h){ - MpegEncContext * const s= (MpegEncContext *)c; +static int dct264_sad8x8_c(/* MpegEncContext */ void *c, uint8_t *src1, + uint8_t *src2, int stride, int h) +{ + MpegEncContext *const s = (MpegEncContext *) c; int16_t dct[8][8]; int i; - int sum=0; + int sum = 0; s->dsp.diff_pixels(dct[0], src1, src2, stride); #define SRC(x) dct[i][x] -#define DST(x,v) dct[i][x]= v - for( i = 0; i < 8; i++ ) +#define DST(x, v) dct[i][x] = v + for (i = 0; i < 8; i++) DCT8_1D #undef SRC #undef DST #define SRC(x) dct[x][i] -#define DST(x,v) sum += FFABS(v) - for( i = 0; i < 8; i++ ) - DCT8_1D +#define DST(x, v) sum += FFABS(v) + for (i = 0; i < 8; i++) + DCT8_1D #undef SRC #undef DST - return sum; + return sum; } #endif -static int dct_max8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *src2, int stride, int h){ - MpegEncContext * const s= (MpegEncContext *)c; +static int dct_max8x8_c(/* MpegEncContext */ void *c, uint8_t *src1, + uint8_t *src2, int stride, int h) +{ + MpegEncContext *const s = (MpegEncContext *) c; LOCAL_ALIGNED_16(int16_t, temp, [64]); - int sum=0, i; + int sum = 0, i; - assert(h==8); + assert(h == 8); s->dsp.diff_pixels(temp, src1, src2, stride); s->dsp.fdct(temp); - for(i=0; i<64; i++) - sum= FFMAX(sum, FFABS(temp[i])); + for (i = 0; i < 64; i++) + sum = FFMAX(sum, FFABS(temp[i])); return sum; } -static int quant_psnr8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *src2, int stride, int h){ - MpegEncContext * const s= (MpegEncContext *)c; - LOCAL_ALIGNED_16(int16_t, temp, [64*2]); - int16_t * const bak = temp+64; - int sum=0, i; +static int quant_psnr8x8_c(/* MpegEncContext */ void *c, uint8_t *src1, + uint8_t *src2, int stride, int h) +{ + MpegEncContext *const s = c; + LOCAL_ALIGNED_16(int16_t, temp, [64 * 2]); + int16_t *const bak = temp + 64; + int sum = 0, i; - assert(h==8); - s->mb_intra=0; + assert(h == 8); + s->mb_intra = 0; s->dsp.diff_pixels(temp, src1, src2, stride); - memcpy(bak, temp, 64*sizeof(int16_t)); + memcpy(bak, temp, 64 * sizeof(int16_t)); - s->block_last_index[0/*FIXME*/]= s->fast_dct_quantize(s, temp, 0/*FIXME*/, s->qscale, &i); + s->block_last_index[0 /* FIXME */] = + s->fast_dct_quantize(s, temp, 0 /* FIXME */, s->qscale, &i); s->dct_unquantize_inter(s, temp, 0, s->qscale); - ff_simple_idct_8(temp); //FIXME + ff_simple_idct_8(temp); // FIXME - for(i=0; i<64; i++) - sum+= (temp[i]-bak[i])*(temp[i]-bak[i]); + for (i = 0; i < 64; i++) + sum += (temp[i] - bak[i]) * (temp[i] - bak[i]); return sum; } -static int rd8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *src2, int stride, int h){ - MpegEncContext * const s= (MpegEncContext *)c; - const uint8_t *scantable= s->intra_scantable.permutated; +static int rd8x8_c(/* MpegEncContext */ void *c, uint8_t *src1, uint8_t *src2, + int stride, int h) +{ + MpegEncContext *const s = (MpegEncContext *) c; + const uint8_t *scantable = s->intra_scantable.permutated; LOCAL_ALIGNED_16(int16_t, temp, [64]); LOCAL_ALIGNED_16(uint8_t, lsrc1, [64]); LOCAL_ALIGNED_16(uint8_t, lsrc2, [64]); int i, last, run, bits, level, distortion, start_i; - const int esc_length= s->ac_esc_length; - uint8_t * length; - uint8_t * last_length; + const int esc_length = s->ac_esc_length; + uint8_t *length; + uint8_t *last_length; - assert(h==8); + assert(h == 8); copy_block8(lsrc1, src1, 8, stride, 8); copy_block8(lsrc2, src2, 8, stride, 8); s->dsp.diff_pixels(temp, lsrc1, lsrc2, 8); - s->block_last_index[0/*FIXME*/]= last= s->fast_dct_quantize(s, temp, 0/*FIXME*/, s->qscale, &i); + s->block_last_index[0 /* FIXME */] = + last = + s->fast_dct_quantize(s, temp, 0 /* FIXME */, s->qscale, &i); - bits=0; + bits = 0; if (s->mb_intra) { - start_i = 1; - length = s->intra_ac_vlc_length; - last_length= s->intra_ac_vlc_last_length; - bits+= s->luma_dc_vlc_length[temp[0] + 256]; //FIXME chroma + start_i = 1; + length = s->intra_ac_vlc_length; + last_length = s->intra_ac_vlc_last_length; + bits += s->luma_dc_vlc_length[temp[0] + 256]; // FIXME: chroma } else { - start_i = 0; - length = s->inter_ac_vlc_length; - last_length= s->inter_ac_vlc_last_length; + start_i = 0; + length = s->inter_ac_vlc_length; + last_length = s->inter_ac_vlc_last_length; } - if(last>=start_i){ - run=0; - for(i=start_i; i= start_i) { + run = 0; + for (i = start_i; i < last; i++) { + int j = scantable[i]; + level = temp[j]; - if(level){ - level+=64; - if((level&(~127)) == 0){ - bits+= length[UNI_AC_ENC_INDEX(run, level)]; - }else - bits+= esc_length; - run=0; - }else + if (level) { + level += 64; + if ((level & (~127)) == 0) + bits += length[UNI_AC_ENC_INDEX(run, level)]; + else + bits += esc_length; + run = 0; + } else run++; } - i= scantable[last]; + i = scantable[last]; - level= temp[i] + 64; + level = temp[i] + 64; assert(level - 64); - if((level&(~127)) == 0){ - bits+= last_length[UNI_AC_ENC_INDEX(run, level)]; - }else - bits+= esc_length; - + if ((level & (~127)) == 0) { + bits += last_length[UNI_AC_ENC_INDEX(run, level)]; + } else + bits += esc_length; } - if(last>=0){ - if(s->mb_intra) + if (last >= 0) { + if (s->mb_intra) s->dct_unquantize_intra(s, temp, 0, s->qscale); else s->dct_unquantize_inter(s, temp, 0, s->qscale); @@ -2148,158 +2448,180 @@ static int rd8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *src2, int s->dsp.idct_add(lsrc2, 8, temp); - distortion= s->dsp.sse[1](NULL, lsrc2, lsrc1, 8, 8); + distortion = s->dsp.sse[1](NULL, lsrc2, lsrc1, 8, 8); - return distortion + ((bits*s->qscale*s->qscale*109 + 64)>>7); + return distortion + ((bits * s->qscale * s->qscale * 109 + 64) >> 7); } -static int bit8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *src2, int stride, int h){ - MpegEncContext * const s= (MpegEncContext *)c; - const uint8_t *scantable= s->intra_scantable.permutated; +static int bit8x8_c(/* MpegEncContext */ void *c, uint8_t *src1, uint8_t *src2, + int stride, int h) +{ + MpegEncContext *const s = (MpegEncContext *) c; + const uint8_t *scantable = s->intra_scantable.permutated; LOCAL_ALIGNED_16(int16_t, temp, [64]); int i, last, run, bits, level, start_i; - const int esc_length= s->ac_esc_length; - uint8_t * length; - uint8_t * last_length; + const int esc_length = s->ac_esc_length; + uint8_t *length; + uint8_t *last_length; - assert(h==8); + assert(h == 8); s->dsp.diff_pixels(temp, src1, src2, stride); - s->block_last_index[0/*FIXME*/]= last= s->fast_dct_quantize(s, temp, 0/*FIXME*/, s->qscale, &i); + s->block_last_index[0 /* FIXME */] = + last = + s->fast_dct_quantize(s, temp, 0 /* FIXME */, s->qscale, &i); - bits=0; + bits = 0; if (s->mb_intra) { - start_i = 1; - length = s->intra_ac_vlc_length; - last_length= s->intra_ac_vlc_last_length; - bits+= s->luma_dc_vlc_length[temp[0] + 256]; //FIXME chroma + start_i = 1; + length = s->intra_ac_vlc_length; + last_length = s->intra_ac_vlc_last_length; + bits += s->luma_dc_vlc_length[temp[0] + 256]; // FIXME: chroma } else { - start_i = 0; - length = s->inter_ac_vlc_length; - last_length= s->inter_ac_vlc_last_length; + start_i = 0; + length = s->inter_ac_vlc_length; + last_length = s->inter_ac_vlc_last_length; } - if(last>=start_i){ - run=0; - for(i=start_i; i= start_i) { + run = 0; + for (i = start_i; i < last; i++) { + int j = scantable[i]; + level = temp[j]; - if(level){ - level+=64; - if((level&(~127)) == 0){ - bits+= length[UNI_AC_ENC_INDEX(run, level)]; - }else - bits+= esc_length; - run=0; - }else + if (level) { + level += 64; + if ((level & (~127)) == 0) + bits += length[UNI_AC_ENC_INDEX(run, level)]; + else + bits += esc_length; + run = 0; + } else run++; } - i= scantable[last]; + i = scantable[last]; - level= temp[i] + 64; + level = temp[i] + 64; assert(level - 64); - if((level&(~127)) == 0){ - bits+= last_length[UNI_AC_ENC_INDEX(run, level)]; - }else - bits+= esc_length; + if ((level & (~127)) == 0) + bits += last_length[UNI_AC_ENC_INDEX(run, level)]; + else + bits += esc_length; } return bits; } -#define VSAD_INTRA(size) \ -static int vsad_intra##size##_c(/*MpegEncContext*/ void *c, uint8_t *s, uint8_t *dummy, int stride, int h){ \ - int score=0; \ - int x,y; \ - \ - for(y=1; y mini) return mini; - else if((a^(1U<<31)) > maxisign) return maxi; - else return a; + if (a > mini) + return mini; + else if ((a ^ (1U << 31)) > maxisign) + return maxi; + else + return a; } -static void vector_clipf_c_opposite_sign(float *dst, const float *src, float *min, float *max, int len){ +static void vector_clipf_c_opposite_sign(float *dst, const float *src, + float *min, float *max, int len) +{ int i; - uint32_t mini = *(uint32_t*)min; - uint32_t maxi = *(uint32_t*)max; - uint32_t maxisign = maxi ^ (1U<<31); - uint32_t *dsti = (uint32_t*)dst; - const uint32_t *srci = (const uint32_t*)src; - for(i=0; i 0) { + + if (min < 0 && max > 0) { vector_clipf_c_opposite_sign(dst, src, &min, &max, len); } else { - for(i=0; i < len; i+=8) { - dst[i ] = av_clipf(src[i ], min, max); + for (i = 0; i < len; i += 8) { + dst[i] = av_clipf(src[i], min, max); dst[i + 1] = av_clipf(src[i + 1], min, max); dst[i + 2] = av_clipf(src[i + 2], min, max); dst[i + 3] = av_clipf(src[i + 3], min, max); @@ -2358,19 +2689,23 @@ static void vector_clipf_c(float *dst, const float *src, float min, float max, i } } -static int32_t scalarproduct_int16_c(const int16_t * v1, const int16_t * v2, int order) +static int32_t scalarproduct_int16_c(const int16_t *v1, const int16_t *v2, + int order) { int res = 0; while (order--) - res += *v1++ * *v2++; + res += *v1++ **v2++; return res; } -static int32_t scalarproduct_and_madd_int16_c(int16_t *v1, const int16_t *v2, const int16_t *v3, int order, int mul) +static int32_t scalarproduct_and_madd_int16_c(int16_t *v1, const int16_t *v2, + const int16_t *v3, + int order, int mul) { int res = 0; + while (order--) { res += *v1 * *v2++; *v1++ += mul * *v3++; @@ -2390,18 +2725,19 @@ static void vector_clip_int32_c(int32_t *dst, const int32_t *src, int32_t min, *dst++ = av_clip(*src++, min, max); *dst++ = av_clip(*src++, min, max); *dst++ = av_clip(*src++, min, max); - len -= 8; + len -= 8; } while (len > 0); } static void jref_idct_put(uint8_t *dest, int line_size, int16_t *block) { - ff_j_rev_dct (block); + ff_j_rev_dct(block); put_pixels_clamped_c(block, dest, line_size); } + static void jref_idct_add(uint8_t *dest, int line_size, int16_t *block) { - ff_j_rev_dct (block); + ff_j_rev_dct(block); add_pixels_clamped_c(block, dest, line_size); } @@ -2410,32 +2746,29 @@ av_cold void ff_dsputil_static_init(void) { int i; - for(i=0;i<512;i++) { + for (i = 0; i < 512; i++) ff_squareTbl[i] = (i - 256) * (i - 256); - } } -av_cold void ff_dsputil_init(DSPContext* c, AVCodecContext *avctx) +av_cold void ff_dsputil_init(DSPContext *c, AVCodecContext *avctx) { #if CONFIG_ENCODERS if (avctx->bits_per_raw_sample == 10) { c->fdct = ff_jpeg_fdct_islow_10; c->fdct248 = ff_fdct248_islow_10; } else { - if(avctx->dct_algo==FF_DCT_FASTINT) { + if (avctx->dct_algo == FF_DCT_FASTINT) { c->fdct = ff_fdct_ifast; c->fdct248 = ff_fdct_ifast248; - } - else if(avctx->dct_algo==FF_DCT_FAAN) { + } else if (avctx->dct_algo == FF_DCT_FAAN) { c->fdct = ff_faandct; c->fdct248 = ff_faandct248; - } - else { - c->fdct = ff_jpeg_fdct_islow_8; //slow/accurate/default + } else { + c->fdct = ff_jpeg_fdct_islow_8; // slow/accurate/default c->fdct248 = ff_fdct248_islow_8; } } -#endif //CONFIG_ENCODERS +#endif /* CONFIG_ENCODERS */ if (avctx->bits_per_raw_sample == 10) { c->idct_put = ff_simple_idct_put_10; @@ -2443,32 +2776,36 @@ av_cold void ff_dsputil_init(DSPContext* c, AVCodecContext *avctx) c->idct = ff_simple_idct_10; c->idct_permutation_type = FF_NO_IDCT_PERM; } else { - if(avctx->idct_algo==FF_IDCT_INT){ - c->idct_put= jref_idct_put; - c->idct_add= jref_idct_add; - c->idct = ff_j_rev_dct; - c->idct_permutation_type= FF_LIBMPEG2_IDCT_PERM; - }else if(avctx->idct_algo==FF_IDCT_FAAN){ - c->idct_put= ff_faanidct_put; - c->idct_add= ff_faanidct_add; - c->idct = ff_faanidct; - c->idct_permutation_type= FF_NO_IDCT_PERM; - }else{ //accurate/default - c->idct_put = ff_simple_idct_put_8; - c->idct_add = ff_simple_idct_add_8; - c->idct = ff_simple_idct_8; - c->idct_permutation_type= FF_NO_IDCT_PERM; + if (avctx->idct_algo == FF_IDCT_INT) { + c->idct_put = jref_idct_put; + c->idct_add = jref_idct_add; + c->idct = ff_j_rev_dct; + c->idct_permutation_type = FF_LIBMPEG2_IDCT_PERM; + } else if (avctx->idct_algo == FF_IDCT_FAAN) { + c->idct_put = ff_faanidct_put; + c->idct_add = ff_faanidct_add; + c->idct = ff_faanidct; + c->idct_permutation_type = FF_NO_IDCT_PERM; + } else { // accurate/default + c->idct_put = ff_simple_idct_put_8; + c->idct_add = ff_simple_idct_add_8; + c->idct = ff_simple_idct_8; + c->idct_permutation_type = FF_NO_IDCT_PERM; } } c->diff_pixels = diff_pixels_c; - c->put_pixels_clamped = put_pixels_clamped_c; + + c->put_pixels_clamped = put_pixels_clamped_c; c->put_signed_pixels_clamped = put_signed_pixels_clamped_c; - c->add_pixels_clamped = add_pixels_clamped_c; + c->add_pixels_clamped = add_pixels_clamped_c; + c->sum_abs_dctelem = sum_abs_dctelem_c; + c->gmc1 = gmc1_c; - c->gmc = ff_gmc_c; - c->pix_sum = pix_sum_c; + c->gmc = ff_gmc_c; + + c->pix_sum = pix_sum_c; c->pix_norm1 = pix_norm1_c; c->fill_block_tab[0] = fill_block16_c; @@ -2484,37 +2821,37 @@ av_cold void ff_dsputil_init(DSPContext* c, AVCodecContext *avctx) c->pix_abs[1][2] = pix_abs8_y2_c; c->pix_abs[1][3] = pix_abs8_xy2_c; - c->put_tpel_pixels_tab[ 0] = put_tpel_pixels_mc00_c; - c->put_tpel_pixels_tab[ 1] = put_tpel_pixels_mc10_c; - c->put_tpel_pixels_tab[ 2] = put_tpel_pixels_mc20_c; - c->put_tpel_pixels_tab[ 4] = put_tpel_pixels_mc01_c; - c->put_tpel_pixels_tab[ 5] = put_tpel_pixels_mc11_c; - c->put_tpel_pixels_tab[ 6] = put_tpel_pixels_mc21_c; - c->put_tpel_pixels_tab[ 8] = put_tpel_pixels_mc02_c; - c->put_tpel_pixels_tab[ 9] = put_tpel_pixels_mc12_c; + c->put_tpel_pixels_tab[0] = put_tpel_pixels_mc00_c; + c->put_tpel_pixels_tab[1] = put_tpel_pixels_mc10_c; + c->put_tpel_pixels_tab[2] = put_tpel_pixels_mc20_c; + c->put_tpel_pixels_tab[4] = put_tpel_pixels_mc01_c; + c->put_tpel_pixels_tab[5] = put_tpel_pixels_mc11_c; + c->put_tpel_pixels_tab[6] = put_tpel_pixels_mc21_c; + c->put_tpel_pixels_tab[8] = put_tpel_pixels_mc02_c; + c->put_tpel_pixels_tab[9] = put_tpel_pixels_mc12_c; c->put_tpel_pixels_tab[10] = put_tpel_pixels_mc22_c; - c->avg_tpel_pixels_tab[ 0] = avg_tpel_pixels_mc00_c; - c->avg_tpel_pixels_tab[ 1] = avg_tpel_pixels_mc10_c; - c->avg_tpel_pixels_tab[ 2] = avg_tpel_pixels_mc20_c; - c->avg_tpel_pixels_tab[ 4] = avg_tpel_pixels_mc01_c; - c->avg_tpel_pixels_tab[ 5] = avg_tpel_pixels_mc11_c; - c->avg_tpel_pixels_tab[ 6] = avg_tpel_pixels_mc21_c; - c->avg_tpel_pixels_tab[ 8] = avg_tpel_pixels_mc02_c; - c->avg_tpel_pixels_tab[ 9] = avg_tpel_pixels_mc12_c; + c->avg_tpel_pixels_tab[0] = avg_tpel_pixels_mc00_c; + c->avg_tpel_pixels_tab[1] = avg_tpel_pixels_mc10_c; + c->avg_tpel_pixels_tab[2] = avg_tpel_pixels_mc20_c; + c->avg_tpel_pixels_tab[4] = avg_tpel_pixels_mc01_c; + c->avg_tpel_pixels_tab[5] = avg_tpel_pixels_mc11_c; + c->avg_tpel_pixels_tab[6] = avg_tpel_pixels_mc21_c; + c->avg_tpel_pixels_tab[8] = avg_tpel_pixels_mc02_c; + c->avg_tpel_pixels_tab[9] = avg_tpel_pixels_mc12_c; c->avg_tpel_pixels_tab[10] = avg_tpel_pixels_mc22_c; -#define dspfunc(PFX, IDX, NUM) \ - c->PFX ## _pixels_tab[IDX][ 0] = PFX ## NUM ## _mc00_c; \ - c->PFX ## _pixels_tab[IDX][ 1] = PFX ## NUM ## _mc10_c; \ - c->PFX ## _pixels_tab[IDX][ 2] = PFX ## NUM ## _mc20_c; \ - c->PFX ## _pixels_tab[IDX][ 3] = PFX ## NUM ## _mc30_c; \ - c->PFX ## _pixels_tab[IDX][ 4] = PFX ## NUM ## _mc01_c; \ - c->PFX ## _pixels_tab[IDX][ 5] = PFX ## NUM ## _mc11_c; \ - c->PFX ## _pixels_tab[IDX][ 6] = PFX ## NUM ## _mc21_c; \ - c->PFX ## _pixels_tab[IDX][ 7] = PFX ## NUM ## _mc31_c; \ - c->PFX ## _pixels_tab[IDX][ 8] = PFX ## NUM ## _mc02_c; \ - c->PFX ## _pixels_tab[IDX][ 9] = PFX ## NUM ## _mc12_c; \ +#define dspfunc(PFX, IDX, NUM) \ + c->PFX ## _pixels_tab[IDX][0] = PFX ## NUM ## _mc00_c; \ + c->PFX ## _pixels_tab[IDX][1] = PFX ## NUM ## _mc10_c; \ + c->PFX ## _pixels_tab[IDX][2] = PFX ## NUM ## _mc20_c; \ + c->PFX ## _pixels_tab[IDX][3] = PFX ## NUM ## _mc30_c; \ + c->PFX ## _pixels_tab[IDX][4] = PFX ## NUM ## _mc01_c; \ + c->PFX ## _pixels_tab[IDX][5] = PFX ## NUM ## _mc11_c; \ + c->PFX ## _pixels_tab[IDX][6] = PFX ## NUM ## _mc21_c; \ + c->PFX ## _pixels_tab[IDX][7] = PFX ## NUM ## _mc31_c; \ + c->PFX ## _pixels_tab[IDX][8] = PFX ## NUM ## _mc02_c; \ + c->PFX ## _pixels_tab[IDX][9] = PFX ## NUM ## _mc12_c; \ c->PFX ## _pixels_tab[IDX][10] = PFX ## NUM ## _mc22_c; \ c->PFX ## _pixels_tab[IDX][11] = PFX ## NUM ## _mc32_c; \ c->PFX ## _pixels_tab[IDX][12] = PFX ## NUM ## _mc03_c; \ @@ -2523,92 +2860,95 @@ av_cold void ff_dsputil_init(DSPContext* c, AVCodecContext *avctx) c->PFX ## _pixels_tab[IDX][15] = PFX ## NUM ## _mc33_c dspfunc(put_qpel, 0, 16); - dspfunc(put_no_rnd_qpel, 0, 16); - - dspfunc(avg_qpel, 0, 16); - dspfunc(put_qpel, 1, 8); + + dspfunc(put_no_rnd_qpel, 0, 16); dspfunc(put_no_rnd_qpel, 1, 8); + dspfunc(avg_qpel, 0, 16); dspfunc(avg_qpel, 1, 8); #undef dspfunc - c->put_mspel_pixels_tab[0]= ff_put_pixels8x8_c; - c->put_mspel_pixels_tab[1]= put_mspel8_mc10_c; - c->put_mspel_pixels_tab[2]= put_mspel8_mc20_c; - c->put_mspel_pixels_tab[3]= put_mspel8_mc30_c; - c->put_mspel_pixels_tab[4]= put_mspel8_mc02_c; - c->put_mspel_pixels_tab[5]= put_mspel8_mc12_c; - c->put_mspel_pixels_tab[6]= put_mspel8_mc22_c; - c->put_mspel_pixels_tab[7]= put_mspel8_mc32_c; + c->put_mspel_pixels_tab[0] = ff_put_pixels8x8_c; + c->put_mspel_pixels_tab[1] = put_mspel8_mc10_c; + c->put_mspel_pixels_tab[2] = put_mspel8_mc20_c; + c->put_mspel_pixels_tab[3] = put_mspel8_mc30_c; + c->put_mspel_pixels_tab[4] = put_mspel8_mc02_c; + c->put_mspel_pixels_tab[5] = put_mspel8_mc12_c; + c->put_mspel_pixels_tab[6] = put_mspel8_mc22_c; + c->put_mspel_pixels_tab[7] = put_mspel8_mc32_c; -#define SET_CMP_FUNC(name) \ - c->name[0]= name ## 16_c;\ - c->name[1]= name ## 8x8_c; +#define SET_CMP_FUNC(name) \ + c->name[0] = name ## 16_c; \ + c->name[1] = name ## 8x8_c; SET_CMP_FUNC(hadamard8_diff) - c->hadamard8_diff[4]= hadamard8_intra16_c; - c->hadamard8_diff[5]= hadamard8_intra8x8_c; + c->hadamard8_diff[4] = hadamard8_intra16_c; + c->hadamard8_diff[5] = hadamard8_intra8x8_c; SET_CMP_FUNC(dct_sad) SET_CMP_FUNC(dct_max) #if CONFIG_GPL SET_CMP_FUNC(dct264_sad) #endif - c->sad[0]= pix_abs16_c; - c->sad[1]= pix_abs8_c; - c->sse[0]= sse16_c; - c->sse[1]= sse8_c; - c->sse[2]= sse4_c; + c->sad[0] = pix_abs16_c; + c->sad[1] = pix_abs8_c; + c->sse[0] = sse16_c; + c->sse[1] = sse8_c; + c->sse[2] = sse4_c; SET_CMP_FUNC(quant_psnr) SET_CMP_FUNC(rd) SET_CMP_FUNC(bit) - c->vsad[0]= vsad16_c; - c->vsad[4]= vsad_intra16_c; - c->vsad[5]= vsad_intra8_c; - c->vsse[0]= vsse16_c; - c->vsse[4]= vsse_intra16_c; - c->vsse[5]= vsse_intra8_c; - c->nsse[0]= nsse16_c; - c->nsse[1]= nsse8_c; + c->vsad[0] = vsad16_c; + c->vsad[4] = vsad_intra16_c; + c->vsad[5] = vsad_intra8_c; + c->vsse[0] = vsse16_c; + c->vsse[4] = vsse_intra16_c; + c->vsse[5] = vsse_intra8_c; + c->nsse[0] = nsse16_c; + c->nsse[1] = nsse8_c; c->ssd_int8_vs_int16 = ssd_int8_vs_int16_c; - c->add_bytes= add_bytes_c; - c->diff_bytes= diff_bytes_c; - c->add_hfyu_median_prediction= add_hfyu_median_prediction_c; - c->sub_hfyu_median_prediction= sub_hfyu_median_prediction_c; - c->add_hfyu_left_prediction = add_hfyu_left_prediction_c; + c->add_bytes = add_bytes_c; + c->add_hfyu_median_prediction = add_hfyu_median_prediction_c; + c->add_hfyu_left_prediction = add_hfyu_left_prediction_c; c->add_hfyu_left_prediction_bgr32 = add_hfyu_left_prediction_bgr32_c; - c->bswap_buf= bswap_buf; + + c->diff_bytes = diff_bytes_c; + c->sub_hfyu_median_prediction = sub_hfyu_median_prediction_c; + + c->bswap_buf = bswap_buf; c->bswap16_buf = bswap16_buf; - c->try_8x8basis= try_8x8basis_c; - c->add_8x8basis= add_8x8basis_c; + c->try_8x8basis = try_8x8basis_c; + c->add_8x8basis = add_8x8basis_c; - c->vector_clipf = vector_clipf_c; - c->scalarproduct_int16 = scalarproduct_int16_c; c->scalarproduct_and_madd_int16 = scalarproduct_and_madd_int16_c; - c->vector_clip_int32 = vector_clip_int32_c; - c->shrink[0]= av_image_copy_plane; - c->shrink[1]= ff_shrink22; - c->shrink[2]= ff_shrink44; - c->shrink[3]= ff_shrink88; + c->scalarproduct_int16 = scalarproduct_int16_c; + c->vector_clip_int32 = vector_clip_int32_c; + c->vector_clipf = vector_clipf_c; + + c->shrink[0] = av_image_copy_plane; + c->shrink[1] = ff_shrink22; + c->shrink[2] = ff_shrink44; + c->shrink[3] = ff_shrink88; c->add_pixels8 = add_pixels8_c; #undef FUNC #undef FUNCC -#define FUNC(f, depth) f ## _ ## depth +#define FUNC(f, depth) f ## _ ## depth #define FUNCC(f, depth) f ## _ ## depth ## _c - c->draw_edges = FUNCC(draw_edges, 8); - c->clear_block = FUNCC(clear_block, 8); - c->clear_blocks = FUNCC(clear_blocks, 8); + c->draw_edges = FUNCC(draw_edges, 8); -#define BIT_DEPTH_FUNCS(depth) \ - c->get_pixels = FUNCC(get_pixels, depth); + c->clear_block = FUNCC(clear_block, 8); + c->clear_blocks = FUNCC(clear_blocks, 8); + +#define BIT_DEPTH_FUNCS(depth) \ + c->get_pixels = FUNCC(get_pixels, depth); switch (avctx->bits_per_raw_sample) { case 9: @@ -2620,7 +2960,6 @@ av_cold void ff_dsputil_init(DSPContext* c, AVCodecContext *avctx) break; } - if (ARCH_ARM) ff_dsputil_init_arm(c, avctx); if (ARCH_BFIN)