Lines Matching refs:TMP2
204 .macro PRECOMPUTE SUBKEY TMP1 TMP2 TMP3 TMP4 TMP5 TMP6 TMP7
207 movdqa SHUF_MASK(%rip), \TMP2
208 PSHUFB_XMM \TMP2, \TMP3
212 movdqa \TMP3, \TMP2
214 psrlq $63, \TMP2
215 movdqa \TMP2, \TMP1
216 pslldq $8, \TMP2
218 por \TMP2, \TMP3
222 pshufd $0x24, \TMP1, \TMP2
223 pcmpeqd TWOONE(%rip), \TMP2
224 pand POLY(%rip), \TMP2
225 pxor \TMP2, \TMP3
233 GHASH_MUL \TMP5, \TMP3, \TMP1, \TMP2, \TMP4, \TMP6, \TMP7
241 GHASH_MUL \TMP5, \TMP3, \TMP1, \TMP2, \TMP4, \TMP6, \TMP7
248 GHASH_MUL \TMP5, \TMP3, \TMP1, \TMP2, \TMP4, \TMP6, \TMP7
519 .macro GHASH_MUL GH HK TMP1 TMP2 TMP3 TMP4 TMP5
521 pshufd $78, \GH, \TMP2
523 pxor \GH, \TMP2 # TMP2 = a1+a0
527 PCLMULQDQ 0x00, \TMP3, \TMP2 # TMP2 = (a0+a1)*(b1+b0)
528 pxor \GH, \TMP2
529 pxor \TMP1, \TMP2 # TMP2 = (a0*b0)+(a1*b0)
530 movdqa \TMP2, \TMP3
532 psrldq $8, \TMP2 # right shift TMP2 2 DWs
534 pxor \TMP2, \TMP1 # TMP2:GH holds the result of GH*HK
538 movdqa \GH, \TMP2
540 movdqa \GH, \TMP4 # copy GH into TMP2,TMP3 and TMP4
543 pslld $31, \TMP2 # packed right shift <<31
546 pxor \TMP3, \TMP2 # xor the shifted versions
547 pxor \TMP4, \TMP2
548 movdqa \TMP2, \TMP5
550 pslldq $12, \TMP2 # left shift TMP2 3 DWs
551 pxor \TMP2, \GH
555 movdqa \GH,\TMP2 # copy GH into TMP2,TMP3 and TMP4
560 psrld $1,\TMP2 # packed left shift >>1
563 pxor \TMP3,\TMP2 # xor the shifted versions
564 pxor \TMP4,\TMP2
565 pxor \TMP5, \TMP2
566 pxor \TMP2, \GH
603 .macro CALC_AAD_HASH HASHKEY AAD AADLEN TMP1 TMP2 TMP3 TMP4 TMP5 \
617 GHASH_MUL \TMP6, \HASHKEY, \TMP1, \TMP2, \TMP3, \TMP4, \TMP5
633 GHASH_MUL \TMP7, \HASHKEY, \TMP1, \TMP2, \TMP3, \TMP4, \TMP5
796 .macro INITIAL_BLOCKS_ENC_DEC TMP1 TMP2 TMP3 TMP4 TMP5 XMM0 XMM1 \
809 MOVADQ 0(%arg1),\TMP2
818 pxor \TMP2, %xmm\index
858 GHASH_MUL %xmm6, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
860 GHASH_MUL %xmm7, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
862 GHASH_MUL %xmm8, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
865 GHASH_MUL %xmm7, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
867 GHASH_MUL %xmm8, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
870 GHASH_MUL %xmm8, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
923 MOVADQ (%r10),\TMP2
925 AESENC \TMP2, %xmm\index
932 MOVADQ (%r10), \TMP2
933 AESENCLAST \TMP2, \XMM1
934 AESENCLAST \TMP2, \XMM2
935 AESENCLAST \TMP2, \XMM3
936 AESENCLAST \TMP2, \XMM4
985 .macro GHASH_4_ENCRYPT_4_PARALLEL_ENC TMP1 TMP2 TMP3 TMP4 TMP5 \
1032 pshufd $78, \XMM6, \TMP2
1033 pxor \XMM6, \TMP2
1048 PCLMULQDQ 0x00, \TMP5, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1057 pxor \TMP2, \TMP6
1059 pshufd $78, \XMM7, \TMP2
1060 pxor \XMM7, \TMP2
1078 PCLMULQDQ 0x00, \TMP5, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1087 pxor \TMP2, \TMP6
1093 pshufd $78, \XMM8, \TMP2
1094 pxor \XMM8, \TMP2
1125 PCLMULQDQ 0x00, \TMP5, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1145 pxor \TMP6, \TMP2
1146 pxor \TMP1, \TMP2
1147 pxor \XMM5, \TMP2
1148 movdqa \TMP2, \TMP3
1150 psrldq $8, \TMP2 # right shift TMP2 2 DWs
1152 pxor \TMP2, \TMP1 # accumulate the results in TMP1:XMM5
1156 movdqa \XMM5, \TMP2
1159 # move XMM5 into TMP2, TMP3, TMP4 in order to perform shifts independently
1160 pslld $31, \TMP2 # packed right shift << 31
1163 pxor \TMP3, \TMP2 # xor the shifted versions
1164 pxor \TMP4, \TMP2
1165 movdqa \TMP2, \TMP5
1167 pslldq $12, \TMP2 # left shift T2 3 DWs
1168 pxor \TMP2, \XMM5
1172 movdqa \XMM5,\TMP2 # make 3 copies of XMM5 into TMP2, TMP3, TMP4
1175 psrld $1, \TMP2 # packed left shift >>1
1178 pxor \TMP3,\TMP2 # xor the shifted versions
1179 pxor \TMP4,\TMP2
1180 pxor \TMP5, \TMP2
1181 pxor \TMP2, \XMM5
1193 .macro GHASH_4_ENCRYPT_4_PARALLEL_DEC TMP1 TMP2 TMP3 TMP4 TMP5 \
1240 pshufd $78, \XMM6, \TMP2
1241 pxor \XMM6, \TMP2
1256 PCLMULQDQ 0x00, \TMP5, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1265 pxor \TMP2, \TMP6
1267 pshufd $78, \XMM7, \TMP2
1268 pxor \XMM7, \TMP2
1286 PCLMULQDQ 0x00, \TMP5, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1295 pxor \TMP2, \TMP6
1301 pshufd $78, \XMM8, \TMP2
1302 pxor \XMM8, \TMP2
1333 PCLMULQDQ 0x00, \TMP5, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1357 pxor \TMP6, \TMP2
1358 pxor \TMP1, \TMP2
1359 pxor \XMM5, \TMP2
1360 movdqa \TMP2, \TMP3
1362 psrldq $8, \TMP2 # right shift TMP2 2 DWs
1364 pxor \TMP2, \TMP1 # accumulate the results in TMP1:XMM5
1368 movdqa \XMM5, \TMP2
1371 # move XMM5 into TMP2, TMP3, TMP4 in order to perform shifts independently
1372 pslld $31, \TMP2 # packed right shift << 31
1375 pxor \TMP3, \TMP2 # xor the shifted versions
1376 pxor \TMP4, \TMP2
1377 movdqa \TMP2, \TMP5
1379 pslldq $12, \TMP2 # left shift T2 3 DWs
1380 pxor \TMP2, \XMM5
1384 movdqa \XMM5,\TMP2 # make 3 copies of XMM5 into TMP2, TMP3, TMP4
1387 psrld $1, \TMP2 # packed left shift >>1
1390 pxor \TMP3,\TMP2 # xor the shifted versions
1391 pxor \TMP4,\TMP2
1392 pxor \TMP5, \TMP2
1393 pxor \TMP2, \XMM5
1400 .macro GHASH_LAST_4 TMP1 TMP2 TMP3 TMP4 TMP5 TMP6 \
1406 pshufd $78, \XMM1, \TMP2
1407 pxor \XMM1, \TMP2
1412 PCLMULQDQ 0x00, \TMP4, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1414 movdqa \TMP2, \XMM1 # result in TMP6, XMMDst, XMM1
1419 pshufd $78, \XMM2, \TMP2
1420 pxor \XMM2, \TMP2
1425 PCLMULQDQ 0x00, \TMP4, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1428 pxor \TMP2, \XMM1
1434 pshufd $78, \XMM3, \TMP2
1435 pxor \XMM3, \TMP2
1440 PCLMULQDQ 0x00, \TMP4, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1443 pxor \TMP2, \XMM1 # results accumulated in TMP6, XMMDst, XMM1
1447 pshufd $78, \XMM4, \TMP2
1448 pxor \XMM4, \TMP2
1453 PCLMULQDQ 0x00, \TMP4, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1456 pxor \XMM1, \TMP2
1457 pxor \TMP6, \TMP2
1458 pxor \XMMDst, \TMP2
1460 movdqa \TMP2, \TMP4
1462 psrldq $8, \TMP2 # right shift TMP2 2 DWs
1464 pxor \TMP2, \TMP6
1467 movdqa \XMMDst, \TMP2
1470 # move XMMDst into TMP2, TMP3, TMP4 in order to perform 3 shifts independently
1471 pslld $31, \TMP2 # packed right shifting << 31
1474 pxor \TMP3, \TMP2 # xor the shifted versions
1475 pxor \TMP4, \TMP2
1476 movdqa \TMP2, \TMP7
1478 pslldq $12, \TMP2 # left shift TMP2 3 DWs
1479 pxor \TMP2, \XMMDst
1482 movdqa \XMMDst, \TMP2
1486 psrld $1, \TMP2 # packed left shift >> 1
1489 pxor \TMP3, \TMP2 # xor the shifted versions
1490 pxor \TMP4, \TMP2
1491 pxor \TMP7, \TMP2
1492 pxor \TMP2, \XMMDst