Lines Matching refs:BPS
58 LD_SB4(dst, BPS, dest0, dest1, dest2, dest3); in TransformOne()
67 ST4x4_UB(res0, res0, 3, 2, 1, 0, dst, BPS); in TransformOne()
122 ADDBLK_ST4x4_UB(tmp0, tmp0, tmp0, tmp0, dst, BPS); in TransformDC()
144 LD_UB4(dst, BPS, dest0, dest1, dest2, dest3); in TransformAC3()
153 ST4x4_UB(res0, res0, 3, 2, 1, 0, dst, BPS); in TransformAC3()
683 for (i = 0; i < 4; ++i) dc += dst[i - BPS] + dst[-1 + i * BPS]; in DC4()
686 SW4(dc, dc, dc, dc, dst, BPS); in DC4()
690 const uint8_t* const ptemp = dst - BPS - 1; in TM4()
693 const v8i16 TL = (v8i16)__msa_fill_h(ptemp[0 * BPS]); in TM4()
694 const v8i16 L0 = (v8i16)__msa_fill_h(ptemp[1 * BPS]); in TM4()
695 const v8i16 L1 = (v8i16)__msa_fill_h(ptemp[2 * BPS]); in TM4()
696 const v8i16 L2 = (v8i16)__msa_fill_h(ptemp[3 * BPS]); in TM4()
697 const v8i16 L3 = (v8i16)__msa_fill_h(ptemp[4 * BPS]); in TM4()
704 PCKEV_ST4x4_UB(r0, r1, r2, r3, dst, BPS); in TM4()
708 const uint8_t* const ptop = dst - BPS - 1; in VE4()
721 SW4(out, out, out, out, dst, BPS); in VE4()
725 const uint8_t* const ptop = dst - 1 - BPS; in RD4()
733 A = (v16u8)__msa_insert_b((v16i8)A, 3, ptop[1 * BPS]); in RD4()
734 A = (v16u8)__msa_insert_b((v16i8)A, 2, ptop[2 * BPS]); in RD4()
735 A = (v16u8)__msa_insert_b((v16i8)A, 1, ptop[3 * BPS]); in RD4()
736 A = (v16u8)__msa_insert_b((v16i8)A, 0, ptop[4 * BPS]); in RD4()
749 SW4(val0, val1, val2, val3, dst, BPS); in RD4()
753 const uint8_t* const ptop = dst - BPS; in LD4()
773 SW4(val0, val1, val2, val3, dst, BPS); in LD4()
781 const v16u8 rtop = LD_UB(dst - BPS); in DC16()
786 dc += dst[-1 + i * BPS]; in DC16()
790 ST_UB8(out, out, out, out, out, out, out, out, dst, BPS); in DC16()
791 ST_UB8(out, out, out, out, out, out, out, out, dst + 8 * BPS, BPS); in DC16()
798 const v8i16 TL = (v8i16)__msa_fill_h(dst[-1 - BPS]); in TM16()
799 const v16i8 T = LD_SB(dst - BPS); in TM16()
806 const v8i16 L0 = (v8i16)__msa_fill_h(dst[-1 + 0 * BPS]); in TM16()
807 const v8i16 L1 = (v8i16)__msa_fill_h(dst[-1 + 1 * BPS]); in TM16()
808 const v8i16 L2 = (v8i16)__msa_fill_h(dst[-1 + 2 * BPS]); in TM16()
809 const v8i16 L3 = (v8i16)__msa_fill_h(dst[-1 + 3 * BPS]); in TM16()
815 ST_SB4(t0, t1, t2, t3, dst, BPS); in TM16()
816 dst += 4 * BPS; in TM16()
821 const v16u8 rtop = LD_UB(dst - BPS); in VE16()
822 ST_UB8(rtop, rtop, rtop, rtop, rtop, rtop, rtop, rtop, dst, BPS); in VE16()
823 ST_UB8(rtop, rtop, rtop, rtop, rtop, rtop, rtop, rtop, dst + 8 * BPS, BPS); in VE16()
829 const v16u8 L0 = (v16u8)__msa_fill_b(dst[-1 + 0 * BPS]); in HE16()
830 const v16u8 L1 = (v16u8)__msa_fill_b(dst[-1 + 1 * BPS]); in HE16()
831 const v16u8 L2 = (v16u8)__msa_fill_b(dst[-1 + 2 * BPS]); in HE16()
832 const v16u8 L3 = (v16u8)__msa_fill_b(dst[-1 + 3 * BPS]); in HE16()
833 ST_UB4(L0, L1, L2, L3, dst, BPS); in HE16()
834 dst += 4 * BPS; in HE16()
844 dc += dst[-1 + j * BPS]; in DC16NoTop()
847 ST_UB8(out, out, out, out, out, out, out, out, dst, BPS); in DC16NoTop()
848 ST_UB8(out, out, out, out, out, out, out, out, dst + 8 * BPS, BPS); in DC16NoTop()
853 const v16u8 rtop = LD_UB(dst - BPS); in DC16NoLeft()
859 ST_UB8(out, out, out, out, out, out, out, out, dst, BPS); in DC16NoLeft()
860 ST_UB8(out, out, out, out, out, out, out, out, dst + 8 * BPS, BPS); in DC16NoLeft()
865 ST_UB8(out, out, out, out, out, out, out, out, dst, BPS); in DC16NoTopLeft()
866 ST_UB8(out, out, out, out, out, out, out, out, dst + 8 * BPS, BPS); in DC16NoTopLeft()
872 SD4(out, out, out, out, dst + 0 * BPS, BPS); \
873 SD4(out, out, out, out, dst + 4 * BPS, BPS); \
880 const v16u8 rtop = LD_UB(dst - BPS); in DC8uv()
887 dc += dst[-1 + i * BPS]; in DC8uv()
897 const v16i8 T1 = LD_SB(dst - BPS); in TM8uv()
900 const v8i16 TL = (v8i16)__msa_fill_h(dst[-1 - BPS]); in TM8uv()
905 v8i16 r0 = (v8i16)__msa_fill_h(dst[-1 + 0 * BPS]); in TM8uv()
906 v8i16 r1 = (v8i16)__msa_fill_h(dst[-1 + 1 * BPS]); in TM8uv()
907 v8i16 r2 = (v8i16)__msa_fill_h(dst[-1 + 2 * BPS]); in TM8uv()
908 v8i16 r3 = (v8i16)__msa_fill_h(dst[-1 + 3 * BPS]); in TM8uv()
912 ST4x4_UB(t0, t1, 0, 2, 0, 2, dst, BPS); in TM8uv()
913 ST4x4_UB(t0, t1, 1, 3, 1, 3, dst + 4, BPS); in TM8uv()
914 dst += 4 * BPS; in TM8uv()
919 const v16u8 rtop = LD_UB(dst - BPS); in VE8uv()
927 const v16u8 L0 = (v16u8)__msa_fill_b(dst[-1 + 0 * BPS]); in HE8uv()
928 const v16u8 L1 = (v16u8)__msa_fill_b(dst[-1 + 1 * BPS]); in HE8uv()
929 const v16u8 L2 = (v16u8)__msa_fill_b(dst[-1 + 2 * BPS]); in HE8uv()
930 const v16u8 L3 = (v16u8)__msa_fill_b(dst[-1 + 3 * BPS]); in HE8uv()
935 SD4(out0, out1, out2, out3, dst, BPS); in HE8uv()
936 dst += 4 * BPS; in HE8uv()
942 const v16u8 rtop = LD_UB(dst - BPS); in DC8uvNoLeft()
959 dc += dst[-1 + i * BPS]; in DC8uvNoTop()