32 static const vec_s16 trans4[4] = {
33 { 64, 64, 64, 64, 64, 64, 64, 64 },
34 { 83, 36, 83, 36, 83, 36, 83, 36 },
35 { 64, -64, 64, -64, 64, -64, 64, -64 },
36 { 36, -83, 36, -83, 36, -83, 36, -83 },
40 { 0x00, 0x01, 0x08, 0x09, 0x10, 0x11, 0x18, 0x19, 0x02, 0x03, 0x0A, 0x0B, 0x12, 0x13, 0x1A, 0x1B },
41 { 0x04, 0x05, 0x0C, 0x0D, 0x14, 0x15, 0x1C, 0x1D, 0x06, 0x07, 0x0E, 0x0F, 0x16, 0x17, 0x1E, 0x1F },
53 src_13 = vec_mergel(src_01, src_23);
54 src_02 = vec_mergeh(src_01, src_23);
56 e0 = vec_msums(src_02, trans4[0],
zero);
57 o0 = vec_msums(src_13, trans4[1],
zero);
58 e1 = vec_msums(src_02, trans4[2],
zero);
59 o1 = vec_msums(src_13, trans4[3],
zero);
62 case 7: add = vec_sl(vec_splat_s32(1), vec_splat_u32( 7 - 1));
break;
63 case 10: add = vec_sl(vec_splat_s32(1), vec_splat_u32(10 - 1));
break;
64 case 12: add = vec_sl(vec_splat_s32(1), vec_splat_u32(12 - 1));
break;
68 e0 = vec_add(e0, add);
69 e1 = vec_add(e1, add);
71 res[0] = vec_add(e0, o0);
72 res[1] = vec_add(e1, o1);
73 res[2] = vec_sub(e1, o1);
74 res[3] = vec_sub(e0, o0);
84 case 7: v_shift = vec_splat_u32(7) ;
break;
85 case 10: v_shift = vec_splat_u32(10);
break;
86 case 12: v_shift = vec_splat_u32(12);
break;
90 for (
i = 0;
i < 4;
i++)
91 res[
i] = vec_sra(res[
i], v_shift);
94 res_packed[0] = vec_packs(res[0], res[1]);
95 res_packed[1] = vec_packs(res[2], res[3]);
98 #define FUNCDECL(a, depth) a ## _ ## depth ## _altivec
99 #define FUNC(a, b) FUNCDECL(a, b)
117 c->idct[0] = ff_hevc_idct_4x4_8_altivec;
119 c->idct[0] = ff_hevc_idct_4x4_10_altivec;