} ScanTable;
void ff_init_scantable(uint8_t *, ScanTable *st, const uint8_t *src_scantable);
+void ff_init_scantable_permutation(uint8_t *idct_permutation,
+ int idct_permutation_type);
#define EMULATED_EDGE(depth) \
void ff_emulated_edge_mc_ ## depth (uint8_t *buf, const uint8_t *src, int linesize,\
*/
void (*vector_fmul_scalar)(float *dst, const float *src, float mul,
int len);
+ /**
+ * Multiply a vector of floats by a scalar float and add to
+ * destination vector. Source and destination vectors must
+ * overlap exactly or not at all.
+ * @param dst result vector, 16-byte aligned
+ * @param src input vector, 16-byte aligned
+ * @param mul scalar value
+ * @param len length of vector, multiple of 4
+ */
+ void (*vector_fmac_scalar)(float *dst, const float *src, float mul,
+ int len);
/**
* Calculate the scalar product of two vectors of floats.
* @param v1 first vector, 16-byte aligned
*/
void (*butterflies_float)(float *restrict v1, float *restrict v2, int len);
+ /**
+ * Calculate the sum and difference of two vectors of floats and interleave
+ * results into a separate output vector of floats, with each sum
+ * positioned before the corresponding difference.
+ *
+ * @param dst output vector
+ * constraints: 16-byte aligned
+ * @param src0 first input vector
+ * constraints: 32-byte aligned
+ * @param src1 second input vector
+ * constraints: 32-byte aligned
+ * @param len number of elements in the input
+ * constraints: multiple of 8
+ */
+ void (*butterflies_float_interleave)(float *dst, const float *src0,
+ const float *src1, int len);
+
/* (I)DCT */
void (*fdct)(DCTELEM *block/* align 16*/);
void (*fdct248)(DCTELEM *block/* align 16*/);
* with the zigzag/alternate scan<br>
* an example to avoid confusion:
* - (->decode coeffs -> zigzag reorder -> dequant -> reference idct ->...)
- * - (x -> referece dct -> reference idct -> x)
- * - (x -> referece dct -> simple_mmx_perm = idct_permutation -> simple_idct_mmx -> x)
+ * - (x -> reference dct -> reference idct -> x)
+ * - (x -> reference dct -> simple_mmx_perm = idct_permutation -> simple_idct_mmx -> x)
* - (->decode coeffs -> zigzag reorder -> simple_mmx_perm -> dequant -> simple_idct_mmx ->...)
*/
uint8_t idct_permutation[64];