| @@ -2377,18 +2377,6 @@ static void butterflies_float_c(float *restrict v1, float *restrict v2, | |||
| } | |||
| } | |||
| static void butterflies_float_interleave_c(float *dst, const float *src0, | |||
| const float *src1, int len) | |||
| { | |||
| int i; | |||
| for (i = 0; i < len; i++) { | |||
| float f1 = src0[i]; | |||
| float f2 = src1[i]; | |||
| dst[2*i ] = f1 + f2; | |||
| dst[2*i + 1] = f1 - f2; | |||
| } | |||
| } | |||
| float ff_scalarproduct_float_c(const float *v1, const float *v2, int len) | |||
| { | |||
| float p = 0.0; | |||
| @@ -2823,7 +2811,6 @@ av_cold void ff_dsputil_init(DSPContext* c, AVCodecContext *avctx) | |||
| c->vector_clip_int32 = vector_clip_int32_c; | |||
| c->scalarproduct_float = ff_scalarproduct_float_c; | |||
| c->butterflies_float = butterflies_float_c; | |||
| c->butterflies_float_interleave = butterflies_float_interleave_c; | |||
| c->shrink[0]= av_image_copy_plane; | |||
| c->shrink[1]= ff_shrink22; | |||
| @@ -364,23 +364,6 @@ typedef struct DSPContext { | |||
| */ | |||
| 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*/); | |||
| @@ -632,50 +632,6 @@ VECTOR_FMUL_ADD | |||
| INIT_YMM avx | |||
| VECTOR_FMUL_ADD | |||
| ;----------------------------------------------------------------------------- | |||
| ; void ff_butterflies_float_interleave(float *dst, const float *src0, | |||
| ; const float *src1, int len); | |||
| ;----------------------------------------------------------------------------- | |||
| %macro BUTTERFLIES_FLOAT_INTERLEAVE 0 | |||
| cglobal butterflies_float_interleave, 4,4,3, dst, src0, src1, len | |||
| %if ARCH_X86_64 | |||
| movsxd lenq, lend | |||
| %endif | |||
| test lenq, lenq | |||
| jz .end | |||
| shl lenq, 2 | |||
| lea src0q, [src0q + lenq] | |||
| lea src1q, [src1q + lenq] | |||
| lea dstq, [ dstq + 2*lenq] | |||
| neg lenq | |||
| .loop: | |||
| mova m0, [src0q + lenq] | |||
| mova m1, [src1q + lenq] | |||
| subps m2, m0, m1 | |||
| addps m0, m0, m1 | |||
| unpcklps m1, m0, m2 | |||
| unpckhps m0, m0, m2 | |||
| %if cpuflag(avx) | |||
| vextractf128 [dstq + 2*lenq ], m1, 0 | |||
| vextractf128 [dstq + 2*lenq + 16], m0, 0 | |||
| vextractf128 [dstq + 2*lenq + 32], m1, 1 | |||
| vextractf128 [dstq + 2*lenq + 48], m0, 1 | |||
| %else | |||
| mova [dstq + 2*lenq ], m1 | |||
| mova [dstq + 2*lenq + mmsize], m0 | |||
| %endif | |||
| add lenq, mmsize | |||
| jl .loop | |||
| .end: | |||
| REP_RET | |||
| %endmacro | |||
| INIT_XMM sse | |||
| BUTTERFLIES_FLOAT_INTERLEAVE | |||
| INIT_YMM avx | |||
| BUTTERFLIES_FLOAT_INTERLEAVE | |||
| ; %1 = aligned/unaligned | |||
| %macro BSWAP_LOOPS 1 | |||
| mov r3, r2 | |||
| @@ -1867,11 +1867,6 @@ void ff_vector_clip_int32_int_sse2(int32_t *dst, const int32_t *src, | |||
| void ff_vector_clip_int32_sse4 (int32_t *dst, const int32_t *src, | |||
| int32_t min, int32_t max, unsigned int len); | |||
| extern void ff_butterflies_float_interleave_sse(float *dst, const float *src0, | |||
| const float *src1, int len); | |||
| extern void ff_butterflies_float_interleave_avx(float *dst, const float *src0, | |||
| const float *src1, int len); | |||
| #define SET_QPEL_FUNCS(PFX, IDX, SIZE, CPU, PREFIX) \ | |||
| do { \ | |||
| c->PFX ## _pixels_tab[IDX][ 0] = PREFIX ## PFX ## SIZE ## _mc00_ ## CPU; \ | |||
| @@ -2149,7 +2144,6 @@ static void dsputil_init_sse(DSPContext *c, AVCodecContext *avctx, int mm_flags) | |||
| c->vector_fmul_add = ff_vector_fmul_add_sse; | |||
| c->scalarproduct_float = ff_scalarproduct_float_sse; | |||
| c->butterflies_float_interleave = ff_butterflies_float_interleave_sse; | |||
| #endif /* HAVE_YASM */ | |||
| } | |||
| @@ -2300,7 +2294,6 @@ static void dsputil_init_avx(DSPContext *c, AVCodecContext *avctx, int mm_flags) | |||
| c->avg_h264_chroma_pixels_tab[0] = ff_avg_h264_chroma_mc8_10_avx; | |||
| } | |||
| } | |||
| c->butterflies_float_interleave = ff_butterflies_float_interleave_avx; | |||
| c->vector_fmul_reverse = ff_vector_fmul_reverse_avx; | |||
| c->vector_fmul_add = ff_vector_fmul_add_avx; | |||
| #endif /* HAVE_AVX_EXTERNAL */ | |||