Lines Matching refs:in3
235 Arguments : Inputs - in0, in1, in2, in3, pdst, stride
239 Stores word from 'in3' to (pdst + 3 * stride)
241 #define SW4(in0, in1, in2, in3, pdst, stride) \
246 SW(in3, (pdst) + 3 * stride); \
250 Arguments : Inputs - in0, in1, in2, in3, pdst, stride
254 Stores double word from 'in3' to (pdst + 3 * stride)
256 #define SD4(in0, in1, in2, in3, pdst, stride) \
261 SD(in3, (pdst) + 3 * stride); \
369 #define ST_V4(RTYPE, in0, in1, in2, in3, pdst, stride) \
372 ST_V2(RTYPE, in2, in3, (pdst) + 2 * stride, stride); \
379 #define ST_V6(RTYPE, in0, in1, in2, in3, in4, in5, pdst, stride) \
381 ST_V4(RTYPE, in0, in1, in2, in3, (pdst), stride); \
386 #define ST_V8(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, pdst, stride) \
388 ST_V4(RTYPE, in0, in1, in2, in3, (pdst), stride); \
511 #define ST_D8(in0, in1, in2, in3, idx0, idx1, idx2, idx3, \
515 ST_D4(in2, in3, idx4, idx5, idx6, idx7, pdst + 4 * stride, stride) \
520 Arguments : Inputs - in0, in1, in2, in3, in4, in5, in6, in7, pdst, stride
527 #define ST12x8_UB(in0, in1, in2, in3, in4, in5, in6, in7, pdst, stride) \
538 out3_m = __msa_copy_u_d((v2i64) in3, 0); \
547 out11_m = __msa_copy_u_w((v4i32) in3, 2); \
579 Arguments : Inputs - in0, in1, in2, in3,
588 Similar for the pair of 'in2' and 'in3'
590 #define AVER_UB2(RTYPE, in0, in1, in2, in3, out0, out1) \
593 out1 = (RTYPE) __msa_aver_u_b((v16u8) in2, (v16u8) in3); \
597 #define AVER_UB4(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
600 AVER_UB2(RTYPE, in0, in1, in2, in3, out0, out1) \
648 Arguments : Inputs - in0, in1, in2, in3, mask0, mask1
653 Selective byte elements from in2 & in3 are copied to out1 as
656 #define VSHF_B2(RTYPE, in0, in1, in2, in3, mask0, mask1, out0, out1) \
659 out1 = (RTYPE) __msa_vshf_b((v16i8) mask1, (v16i8) in3, (v16i8) in2); \
666 #define VSHF_B3(RTYPE, in0, in1, in2, in3, in4, in5, mask0, mask1, mask2, \
669 VSHF_B2(RTYPE, in0, in1, in2, in3, mask0, mask1, out0, out1); \
684 Arguments : Inputs - in0, in1, in2, in3, mask0, mask1
689 Selective halfword elements from in2 & in3 are copied to out1
692 #define VSHF_H2(RTYPE, in0, in1, in2, in3, mask0, mask1, out0, out1) \
695 out1 = (RTYPE) __msa_vshf_h((v8i16) mask1, (v8i16) in3, (v8i16) in2); \
699 #define VSHF_H3(RTYPE, in0, in1, in2, in3, in4, in5, mask0, mask1, mask2, \
702 VSHF_H2(RTYPE, in0, in1, in2, in3, mask0, mask1, out0, out1); \
708 Arguments : Inputs - in0, in1, in2, in3, mask0, mask1
713 Selective byte elements from in2 & in3 are copied to out1 as
716 #define VSHF_W2(RTYPE, in0, in1, in2, in3, mask0, mask1, out0, out1) \
719 out1 = (RTYPE) __msa_vshf_w((v4i32) mask1, (v4i32) in3, (v4i32) in2); \
908 #define MIN_UH4(RTYPE, in0, in1, in2, in3, min_vec) \
911 MIN_UH2(RTYPE, in2, in3, min_vec); \
947 #define CLIP_SH4_0_255(in0, in1, in2, in3) \
950 CLIP_SH2_0_255(in2, in3); \
953 #define CLIP_SH8_0_255(in0, in1, in2, in3, \
956 CLIP_SH4_0_255(in0, in1, in2, in3); \
978 #define CLIP_SW4_0_255(in0, in1, in2, in3) \
981 CLIP_SW2_0_255(in2, in3); \
984 #define CLIP_SW8_0_255(in0, in1, in2, in3, \
987 CLIP_SW4_0_255(in0, in1, in2, in3); \
1046 #define HADD_SB4(RTYPE, in0, in1, in2, in3, out0, out1, out2, out3) \
1049 HADD_SB2(RTYPE, in2, in3, out2, out3); \
1076 #define HADD_UB4(RTYPE, in0, in1, in2, in3, out0, out1, out2, out3) \
1079 HADD_UB2(RTYPE, in2, in3, out2, out3); \
1101 #define HSUB_UB4(RTYPE, in0, in1, in2, in3, out0, out1, out2, out3) \
1104 HSUB_UB2(RTYPE, in2, in3, out2, out3); \
1134 Arguments : Inputs - in0, in1, in2, in3 (4 input vectors)
1146 #define INSERT_W4(RTYPE, in0, in1, in2, in3, out) \
1151 out = (RTYPE) __msa_insert_w((v4i32) out, 3, in3); \
1175 Arguments : Inputs - in0, in1, in2, in3
1181 elements of 'in3' are interleaved and copied to 'out1'
1183 #define ILVEV_B2(RTYPE, in0, in1, in2, in3, out0, out1) \
1186 out1 = (RTYPE) __msa_ilvev_b((v16i8) in3, (v16i8) in2); \
1194 Arguments : Inputs - in0, in1, in2, in3
1200 elements of 'in3' are interleaved and copied to 'out1'
1202 #define ILVEV_H2(RTYPE, in0, in1, in2, in3, out0, out1) \
1205 out1 = (RTYPE) __msa_ilvev_h((v8i16) in3, (v8i16) in2); \
1212 Arguments : Inputs - in0, in1, in2, in3
1218 elements of 'in3' are interleaved and copied to 'out1'
1220 #define ILVEV_W2(RTYPE, in0, in1, in2, in3, out0, out1) \
1223 out1 = (RTYPE) __msa_ilvev_w((v4i32) in3, (v4i32) in2); \
1231 Arguments : Inputs - in0, in1, in2, in3
1237 elements of 'in3' are interleaved and copied to 'out1'
1239 #define ILVEV_D2(RTYPE, in0, in1, in2, in3, out0, out1) \
1242 out1 = (RTYPE) __msa_ilvev_d((v2i64) in3, (v2i64) in2); \
1249 Arguments : Inputs - in0, in1, in2, in3
1255 elements of in3 are interleaved and copied to out1.
1257 #define ILVL_B2(RTYPE, in0, in1, in2, in3, out0, out1) \
1260 out1 = (RTYPE) __msa_ilvl_b((v16i8) in2, (v16i8) in3); \
1267 #define ILVL_B4(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
1270 ILVL_B2(RTYPE, in0, in1, in2, in3, out0, out1); \
1279 Arguments : Inputs - in0, in1, in2, in3
1285 elements of in3 are interleaved and copied to out1.
1287 #define ILVL_H2(RTYPE, in0, in1, in2, in3, out0, out1) \
1290 out1 = (RTYPE) __msa_ilvl_h((v8i16) in2, (v8i16) in3); \
1295 #define ILVL_H4(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
1298 ILVL_H2(RTYPE, in0, in1, in2, in3, out0, out1); \
1305 Arguments : Inputs - in0, in1, in2, in3
1311 elements of in3 are interleaved and copied to out1.
1313 #define ILVL_W2(RTYPE, in0, in1, in2, in3, out0, out1) \
1316 out1 = (RTYPE) __msa_ilvl_w((v4i32) in2, (v4i32) in3); \
1323 Arguments : Inputs - in0, in1, in2, in3, in4, in5, in6, in7
1329 elements of in3 are interleaved and copied to out1.
1332 #define ILVR_B2(RTYPE, in0, in1, in2, in3, out0, out1) \
1335 out1 = (RTYPE) __msa_ilvr_b((v16i8) in2, (v16i8) in3); \
1343 #define ILVR_B3(RTYPE, in0, in1, in2, in3, in4, in5, out0, out1, out2) \
1345 ILVR_B2(RTYPE, in0, in1, in2, in3, out0, out1); \
1353 #define ILVR_B4(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
1356 ILVR_B2(RTYPE, in0, in1, in2, in3, out0, out1); \
1365 #define ILVR_B8(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
1369 ILVR_B4(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
1378 Arguments : Inputs - in0, in1, in2, in3, in4, in5, in6, in7
1384 halfword elements of in3 are interleaved and copied to out1.
1387 #define ILVR_H2(RTYPE, in0, in1, in2, in3, out0, out1) \
1390 out1 = (RTYPE) __msa_ilvr_h((v8i16) in2, (v8i16) in3); \
1395 #define ILVR_H3(RTYPE, in0, in1, in2, in3, in4, in5, out0, out1, out2) \
1397 ILVR_H2(RTYPE, in0, in1, in2, in3, out0, out1); \
1402 #define ILVR_H4(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
1405 ILVR_H2(RTYPE, in0, in1, in2, in3, out0, out1); \
1411 #define ILVR_W2(RTYPE, in0, in1, in2, in3, out0, out1) \
1414 out1 = (RTYPE) __msa_ilvr_w((v4i32) in2, (v4i32) in3); \
1420 #define ILVR_W4(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
1423 ILVR_W2(RTYPE, in0, in1, in2, in3, out0, out1); \
1430 Arguments : Inputs - in0, in1, in2, in3, in4, in5, in6, in7
1436 double word elements of in3 are interleaved and copied to out1.
1438 #define ILVR_D2(RTYPE, in0, in1, in2, in3, out0, out1) \
1441 out1 = (RTYPE) __msa_ilvr_d((v2i64) in2, (v2i64) in3); \
1447 #define ILVR_D3(RTYPE, in0, in1, in2, in3, in4, in5, out0, out1, out2) \
1449 ILVR_D2(RTYPE, in0, in1, in2, in3, out0, out1); \
1454 #define ILVR_D4(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
1457 ILVR_D2(RTYPE, in0, in1, in2, in3, out0, out1); \
1464 Arguments : Inputs - in0, in1, in2, in3
1470 double word elements of in3 are interleaved and copied to out1.
1472 #define ILVL_D2(RTYPE, in0, in1, in2, in3, out0, out1) \
1475 out1 = (RTYPE) __msa_ilvl_d((v2i64) in2, (v2i64) in3); \
1522 Arguments : Inputs - in0, in1, in2, in3, max_val
1523 Outputs - in0, in1, in2, in3 (in place)
1536 #define MAXI_SH4(RTYPE, in0, in1, in2, in3, max_val) \
1539 MAXI_SH2(RTYPE, in2, in3, max_val); \
1544 #define MAXI_SH8(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, max_val) \
1546 MAXI_SH4(RTYPE, in0, in1, in2, in3, max_val); \
1555 Arguments : Inputs - in0, in1, in2, in3, sat_val
1556 Outputs - in0, in1, in2, in3 (in place)
1570 #define SAT_UH4(RTYPE, in0, in1, in2, in3, sat_val) \
1573 SAT_UH2(RTYPE, in2, in3, sat_val); \
1578 #define SAT_UH8(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, sat_val) \
1580 SAT_UH4(RTYPE, in0, in1, in2, in3, sat_val); \
1589 Arguments : Inputs - in0, in1, in2, in3, sat_val
1590 Outputs - in0, in1, in2, in3 (in place)
1610 #define SAT_SH4(RTYPE, in0, in1, in2, in3, sat_val) \
1613 SAT_SH2(RTYPE, in2, in3, sat_val); \
1620 Arguments : Inputs - in0, in1, in2, in3, sat_val
1621 Outputs - in0, in1, in2, in3 (in place)
1634 #define SAT_SW4(RTYPE, in0, in1, in2, in3, sat_val) \
1637 SAT_SW2(RTYPE, in2, in3, sat_val); \
1704 Arguments : Inputs - in0, in1, in2, in3
1711 out1 & even byte elements of in3 are copied to the right
1714 #define PCKEV_B2(RTYPE, in0, in1, in2, in3, out0, out1) \
1717 out1 = (RTYPE) __msa_pckev_b((v16i8) in2, (v16i8) in3); \
1724 #define PCKEV_B3(RTYPE, in0, in1, in2, in3, in4, in5, out0, out1, out2) \
1726 PCKEV_B2(RTYPE, in0, in1, in2, in3, out0, out1); \
1732 #define PCKEV_B4(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
1735 PCKEV_B2(RTYPE, in0, in1, in2, in3, out0, out1); \
1744 Arguments : Inputs - in0, in1, in2, in3
1751 out1 & even halfword elements of in3 are copied to the right
1754 #define PCKEV_H2(RTYPE, in0, in1, in2, in3, out0, out1) \
1757 out1 = (RTYPE) __msa_pckev_h((v8i16) in2, (v8i16) in3); \
1762 #define PCKEV_H4(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
1765 PCKEV_H2(RTYPE, in0, in1, in2, in3, out0, out1); \
1772 Arguments : Inputs - in0, in1, in2, in3
1779 out1 & even double elements of in3 are copied to the right
1782 #define PCKEV_D2(RTYPE, in0, in1, in2, in3, out0, out1) \
1785 out1 = (RTYPE) __msa_pckev_d((v2i64) in2, (v2i64) in3); \
1791 #define PCKEV_D4(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
1794 PCKEV_D2(RTYPE, in0, in1, in2, in3, out0, out1); \
1808 #define PCKOD_D2(RTYPE, in0, in1, in2, in3, out0, out1) \
1811 out1 = (RTYPE) __msa_pckod_d((v2i64) in2, (v2i64) in3); \
1845 #define XORI_B4_128(RTYPE, in0, in1, in2, in3) \
1848 XORI_B2_128(RTYPE, in2, in3); \
1854 #define XORI_B5_128(RTYPE, in0, in1, in2, in3, in4) \
1857 XORI_B2_128(RTYPE, in3, in4); \
1861 #define XORI_B6_128(RTYPE, in0, in1, in2, in3, in4, in5) \
1863 XORI_B4_128(RTYPE, in0, in1, in2, in3); \
1868 #define XORI_B7_128(RTYPE, in0, in1, in2, in3, in4, in5, in6) \
1870 XORI_B4_128(RTYPE, in0, in1, in2, in3); \
1875 #define XORI_B8_128(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7) \
1877 XORI_B4_128(RTYPE, in0, in1, in2, in3); \
1884 Arguments : Inputs - in0, in1, in2, in3
1892 #define ADDS_SH2(RTYPE, in0, in1, in2, in3, out0, out1) \
1895 out1 = (RTYPE) __msa_adds_s_h((v8i16) in2, (v8i16) in3); \
1899 #define ADDS_SH4(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
1902 ADDS_SH2(RTYPE, in0, in1, in2, in3, out0, out1); \
1909 Arguments : Inputs - in0, in1, in2, in3, shift
1910 Outputs - in0, in1, in2, in3 (in place)
1921 #define SLLI_4V(in0, in1, in2, in3, shift) \
1926 in3 = in3 << shift; \
1931 Arguments : Inputs - in0, in1, in2, in3, shift
1932 Outputs - in0, in1, in2, in3 (in place)
1939 #define SRA_4V(in0, in1, in2, in3, shift) \
1944 in3 = in3 >> shift; \
1948 Arguments : Inputs - in0, in1, in2, in3, shift
1949 Outputs - in0, in1, in2, in3 (in place)
1957 #define SRL_H4(RTYPE, in0, in1, in2, in3, shift) \
1962 in3 = (RTYPE) __msa_srl_h((v8i16) in3, (v8i16) shift); \
1966 #define SRLR_H4(RTYPE, in0, in1, in2, in3, shift) \
1971 in3 = (RTYPE) __msa_srlr_h((v8i16) in3, (v8i16) shift); \
1976 #define SRLR_H8(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, shift) \
1978 SRLR_H4(RTYPE, in0, in1, in2, in3, shift); \
2010 #define SRAR_H4(RTYPE, in0, in1, in2, in3, shift) \
2013 SRAR_H2(RTYPE, in2, in3, shift) \
2036 #define SRAR_W4(RTYPE, in0, in1, in2, in3, shift) \
2039 SRAR_W2(RTYPE, in2, in3, shift) \
2044 Arguments : Inputs - in0, in1, in2, in3, shift
2045 Outputs - in0, in1, in2, in3 (in place)
2061 #define SRARI_H4(RTYPE, in0, in1, in2, in3, shift) \
2064 SRARI_H2(RTYPE, in2, in3, shift); \
2086 #define SRARI_W4(RTYPE, in0, in1, in2, in3, shift) \
2089 SRARI_W2(RTYPE, in2, in3, shift); \
2095 Arguments : Inputs - in0, in1, in2, in3
2101 #define MUL2(in0, in1, in2, in3, out0, out1) \
2104 out1 = in2 * in3; \
2106 #define MUL4(in0, in1, in2, in3, in4, in5, in6, in7, out0, out1, out2, out3) \
2108 MUL2(in0, in1, in2, in3, out0, out1); \
2113 Arguments : Inputs - in0, in1, in2, in3
2118 #define ADD2(in0, in1, in2, in3, out0, out1) \
2121 out1 = in2 + in3; \
2123 #define ADD4(in0, in1, in2, in3, in4, in5, in6, in7, out0, out1, out2, out3) \
2125 ADD2(in0, in1, in2, in3, out0, out1); \
2130 Arguments : Inputs - in0, in1, in2, in3
2135 #define SUB2(in0, in1, in2, in3, out0, out1) \
2138 out1 = in2 - in3; \
2140 #define SUB4(in0, in1, in2, in3, in4, in5, in6, in7, out0, out1, out2, out3) \
2143 out1 = in2 - in3; \
2245 Arguments : Inputs - in0, in1, in2, in3
2249 #define BUTTERFLY_4(in0, in1, in2, in3, out0, out1, out2, out3) \
2251 out0 = in0 + in3; \
2255 out3 = in0 - in3; \
2263 #define BUTTERFLY_8(in0, in1, in2, in3, in4, in5, in6, in7, \
2269 out3 = in3 + in4; \
2271 out4 = in3 - in4; \
2282 #define BUTTERFLY_16(in0, in1, in2, in3, in4, in5, in6, in7, \
2290 out3 = in3 + in12; \
2300 out12 = in3 - in12; \
2307 Arguments : Inputs - in0, in1, in2, in3 (input 4x4 byte block)
2312 #define TRANSPOSE4x4_UB_UB(in0, in1, in2, in3, out0, out1, out2, out3) \
2317 ILVR_D2_SB(in1, in0, in3, in2, s0_m, s1_m); \
2327 Arguments : Inputs - in0, in1, in2, in3 (input 8x4 byte block)
2332 #define TRANSPOSE8x4_UB(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
2339 ILVEV_W2_SB(in2, in6, in3, in7, tmp0_m, tmp1_m); \
2352 Arguments : Inputs - in0, in1, in2, in3, in4, in5, in6, in7
2359 #define TRANSPOSE8x8_UB(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
2366 ILVR_B4_SB(in2, in0, in3, in1, in6, in4, in7, in5, \
2379 Arguments : Inputs - in0, in1, in2, in3, in4, in5, in6, in7,
2385 #define TRANSPOSE16x4_UB_UB(in0, in1, in2, in3, in4, in5, in6, in7, \
2400 ILVEV_W2_SD(in3, in7, in11, in15, tmp0_m, tmp1_m); \
2414 Arguments : Inputs - in0, in1, in2, in3, in4, in5, in6, in7,
2420 #define TRANSPOSE16x8_UB_UB(in0, in1, in2, in3, in4, in5, in6, in7, \
2428 ILVEV_D2_UB(in2, in10, in3, in11, out5, out4); \
2461 Arguments : Inputs - in0, in1, in2, in3
2466 #define TRANSPOSE4x4_SH_SH(in0, in1, in2, in3, out0, out1, out2, out3) \
2470 ILVR_H2_SH(in1, in0, in3, in2, s0_m, s1_m); \
2477 Arguments : Inputs - in0, in1, in2, in3, in4, in5, in6, in7
2482 #define TRANSPOSE8x8_H(RTYPE, in0, in1, in2, in3, in4, in5, in6, in7, \
2493 ILVR_H2_SH(in2, in0, in3, in1, s0_m, s1_m); \
2495 ILVL_H2_SH(in2, in0, in3, in1, s0_m, s1_m); \
2508 Arguments : Inputs - in0, in1, in2, in3
2513 #define TRANSPOSE4x4_SW_SW(in0, in1, in2, in3, out0, out1, out2, out3) \
2518 ILVRL_W2_SW(in3, in2, s2_m, s3_m); \
2528 Arguments : Inputs - in0, in1, in2, in3, in4, in5, in6, in7, pdst, stride
2531 Each byte element from input vector pair 'in2' and 'in3' are
2540 #define AVE_ST8x4_UB(in0, in1, in2, in3, in4, in5, in6, in7, pdst, stride) \
2546 tmp1_m = __msa_ave_u_b((v16u8) in2, (v16u8) in3); \
2559 Arguments : Inputs - in0, in1, in2, in3, in4, in5, in6, in7, pdst, stride
2562 Each byte element from input vector pair 'in2' and 'in3' are
2571 #define AVE_ST16x4_UB(in0, in1, in2, in3, in4, in5, in6, in7, pdst, stride) \
2576 tmp1_m = __msa_ave_u_b((v16u8) in2, (v16u8) in3); \
2585 Arguments : Inputs - in0, in1, in2, in3, in4, in5, in6, in7, pdst, stride
2588 Each byte element from input vector pair 'in2' and 'in3' are
2597 #define AVER_ST8x4_UB(in0, in1, in2, in3, in4, in5, in6, in7, pdst, stride) \
2602 AVER_UB4_UB(in0, in1, in2, in3, in4, in5, in6, in7, \
2614 Arguments : Inputs - in0, in1, in2, in3, in4, in5, in6, in7, pdst, stride
2617 Each byte element from input vector pair 'in2' and 'in3' are
2626 #define AVER_ST16x4_UB(in0, in1, in2, in3, in4, in5, in6, in7, pdst, stride) \
2630 AVER_UB4_UB(in0, in1, in2, in3, in4, in5, in6, in7, \
2638 Arguments : Inputs - in0, in1, in2, in3, in4, in5, in6, in7, pdst, stride
2641 Each byte element from input vector pair 'in2' and 'in3' are
2650 #define AVER_DST_ST8x4_UB(in0, in1, in2, in3, in4, in5, in6, in7, \
2657 AVER_UB4_UB(in0, in1, in2, in3, in4, in5, in6, in7, \
2666 Arguments : Inputs - in0, in1, in2, in3, in4, in5, in6, in7, pdst, stride
2669 Each byte element from input vector pair 'in2' and 'in3' are
2678 #define AVER_DST_ST16x4_UB(in0, in1, in2, in3, in4, in5, in6, in7, \
2685 AVER_UB4_UB(in0, in1, in2, in3, in4, in5, in6, in7, \
2692 Arguments : Inputs - in0, in1, in2, in3, pdst, stride
2696 #define ADDBLK_ST4x4_UB(in0, in1, in2, in3, pdst, stride) \
2705 ILVR_D2_SH(in1, in0, in3, in2, inp0_m, inp1_m) \
2761 Arguments : Inputs - in0, in1, in2, in3, dst0, dst1, pdst, stride
2763 #define CONVERT_UB_AVG_ST8x4_UB(in0, in1, in2, in3, \
2770 tmp1_m = PCKEV_XORI128_UB(in2, in3); \
2778 Arguments : Inputs - in0, in1, in2, in3, pdst, stride
2780 #define PCKEV_ST4x4_UB(in0, in1, in2, in3, pdst, stride) \
2785 PCKEV_B2_SB(in1, in0, in3, in2, tmp0_m, tmp1_m); \