Lines Matching refs:dst
72 vdst_orig = vec_ld(0, dst); \
75 #define vdst_load(d) vdst = vec_vsx_ld(0, dst)
84 vec_ste(va_u32, element, (uint32_t*)dst);
86 static void h264_idct_add_altivec(uint8_t *dst, int16_t *block, int stride)
96 vec_u8 vdst_mask = vec_lvsl(0, dst);
97 int element = ((unsigned long)dst & 0xf) >> 2;
118 dst += stride;
120 dst += stride;
122 dst += stride;
218 static void h264_idct8_add_altivec(uint8_t *dst, int16_t *dct, int stride)
225 GET_2PERM(perm_ldv, perm_stv, dst);
254 ALTIVEC_STORE_SUM_CLIP(&dst[0*stride], idct0, perm_ldv, perm_stv, sel);
255 ALTIVEC_STORE_SUM_CLIP(&dst[1*stride], idct1, perm_ldv, perm_stv, sel);
256 ALTIVEC_STORE_SUM_CLIP(&dst[2*stride], idct2, perm_ldv, perm_stv, sel);
257 ALTIVEC_STORE_SUM_CLIP(&dst[3*stride], idct3, perm_ldv, perm_stv, sel);
258 ALTIVEC_STORE_SUM_CLIP(&dst[4*stride], idct4, perm_ldv, perm_stv, sel);
259 ALTIVEC_STORE_SUM_CLIP(&dst[5*stride], idct5, perm_ldv, perm_stv, sel);
260 ALTIVEC_STORE_SUM_CLIP(&dst[6*stride], idct6, perm_ldv, perm_stv, sel);
261 ALTIVEC_STORE_SUM_CLIP(&dst[7*stride], idct7, perm_ldv, perm_stv, sel);
269 static av_always_inline void h264_idct_dc_add_internal(uint8_t *dst, int16_t *block, int stride, int size)
289 aligner = vec_lvsr(0, dst);
295 v0 = DST_LD(0, dst+0*stride);
296 v1 = DST_LD(0, dst+1*stride);
297 v2 = DST_LD(0, dst+2*stride);
298 v3 = DST_LD(0, dst+3*stride);
310 VEC_ST(v0, 0, dst+0*stride);
311 VEC_ST(v1, 0, dst+1*stride);
312 VEC_ST(v2, 0, dst+2*stride);
313 VEC_ST(v3, 0, dst+3*stride);
315 dst += 4*stride;
319 static void h264_idct_dc_add_altivec(uint8_t *dst, int16_t *block, int stride)
321 h264_idct_dc_add_internal(dst, block, stride, 4);
324 static void h264_idct8_dc_add_altivec(uint8_t *dst, int16_t *block, int stride)
326 h264_idct_dc_add_internal(dst, block, stride, 8);
329 static void h264_idct_add16_altivec(uint8_t *dst, const int *block_offset,
337 if(nnz==1 && block[i*16]) h264_idct_dc_add_altivec(dst + block_offset[i], block + i*16, stride);
338 else h264_idct_add_altivec(dst + block_offset[i], block + i*16, stride);
343 static void h264_idct_add16intra_altivec(uint8_t *dst, const int *block_offset,
349 if(nnzc[ scan8[i] ]) h264_idct_add_altivec(dst + block_offset[i], block + i*16, stride);
350 else if(block[i*16]) h264_idct_dc_add_altivec(dst + block_offset[i], block + i*16, stride);
354 static void h264_idct8_add4_altivec(uint8_t *dst, const int *block_offset,
362 if(nnz==1 && block[i*16]) h264_idct8_dc_add_altivec(dst + block_offset[i], block + i*16, stride);
363 else h264_idct8_add_altivec(dst + block_offset[i], block + i*16, stride);
400 static inline void write16x4(uint8_t *dst, int dst_stride,
404 uint32_t *src_int = (uint32_t *)result, *dst_int = (uint32_t *)dst;
430 /** @brief performs a 6x16 transpose of data in src, and stores it to dst
706 void biweight_h264_W_altivec(uint8_t *dst, uint8_t *src, int stride, int height,
730 dst_aligned = !((unsigned long)dst & 0xf);
734 vdst = vec_ld(0, dst);
766 vec_st(vdst, 0, dst);
768 dst += stride;
779 static void biweight_h264_pixels ## W ## _altivec(uint8_t *dst, uint8_t *src, ptrdiff_t stride, int height, \
782 biweight_h264_W_altivec(dst, src, stride, height, log2_denom, weightd, weights, offset, W); \