Lines Matching refs:TMP2
195 .macro PRECOMPUTE SUBKEY TMP1 TMP2 TMP3 TMP4 TMP5 TMP6 TMP7
198 movdqa SHUF_MASK(%rip), \TMP2
199 pshufb \TMP2, \TMP3
203 movdqa \TMP3, \TMP2
205 psrlq $63, \TMP2
206 movdqa \TMP2, \TMP1
207 pslldq $8, \TMP2
209 por \TMP2, \TMP3
213 pshufd $0x24, \TMP1, \TMP2
214 pcmpeqd TWOONE(%rip), \TMP2
215 pand POLY(%rip), \TMP2
216 pxor \TMP2, \TMP3
224 GHASH_MUL \TMP5, \TMP3, \TMP1, \TMP2, \TMP4, \TMP6, \TMP7
232 GHASH_MUL \TMP5, \TMP3, \TMP1, \TMP2, \TMP4, \TMP6, \TMP7
239 GHASH_MUL \TMP5, \TMP3, \TMP1, \TMP2, \TMP4, \TMP6, \TMP7
510 .macro GHASH_MUL GH HK TMP1 TMP2 TMP3 TMP4 TMP5
512 pshufd $78, \GH, \TMP2
514 pxor \GH, \TMP2 # TMP2 = a1+a0
518 pclmulqdq $0x00, \TMP3, \TMP2 # TMP2 = (a0+a1)*(b1+b0)
519 pxor \GH, \TMP2
520 pxor \TMP1, \TMP2 # TMP2 = (a0*b0)+(a1*b0)
521 movdqa \TMP2, \TMP3
523 psrldq $8, \TMP2 # right shift TMP2 2 DWs
525 pxor \TMP2, \TMP1 # TMP2:GH holds the result of GH*HK
529 movdqa \GH, \TMP2
531 movdqa \GH, \TMP4 # copy GH into TMP2,TMP3 and TMP4
534 pslld $31, \TMP2 # packed right shift <<31
537 pxor \TMP3, \TMP2 # xor the shifted versions
538 pxor \TMP4, \TMP2
539 movdqa \TMP2, \TMP5
541 pslldq $12, \TMP2 # left shift TMP2 3 DWs
542 pxor \TMP2, \GH
546 movdqa \GH,\TMP2 # copy GH into TMP2,TMP3 and TMP4
551 psrld $1,\TMP2 # packed left shift >>1
554 pxor \TMP3,\TMP2 # xor the shifted versions
555 pxor \TMP4,\TMP2
556 pxor \TMP5, \TMP2
557 pxor \TMP2, \GH
594 .macro CALC_AAD_HASH HASHKEY AAD AADLEN TMP1 TMP2 TMP3 TMP4 TMP5 \
608 GHASH_MUL \TMP6, \HASHKEY, \TMP1, \TMP2, \TMP3, \TMP4, \TMP5
624 GHASH_MUL \TMP7, \HASHKEY, \TMP1, \TMP2, \TMP3, \TMP4, \TMP5
787 .macro INITIAL_BLOCKS_ENC_DEC TMP1 TMP2 TMP3 TMP4 TMP5 XMM0 XMM1 \
800 MOVADQ 0(%arg1),\TMP2
809 pxor \TMP2, %xmm\index
849 GHASH_MUL %xmm6, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
851 GHASH_MUL %xmm7, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
853 GHASH_MUL %xmm8, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
856 GHASH_MUL %xmm7, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
858 GHASH_MUL %xmm8, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
861 GHASH_MUL %xmm8, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
914 MOVADQ (%r10),\TMP2
916 aesenc \TMP2, %xmm\index
923 MOVADQ (%r10), \TMP2
924 aesenclast \TMP2, \XMM1
925 aesenclast \TMP2, \XMM2
926 aesenclast \TMP2, \XMM3
927 aesenclast \TMP2, \XMM4
976 .macro GHASH_4_ENCRYPT_4_PARALLEL_enc TMP1 TMP2 TMP3 TMP4 TMP5 \
1023 pshufd $78, \XMM6, \TMP2
1024 pxor \XMM6, \TMP2
1039 pclmulqdq $0x00, \TMP5, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1048 pxor \TMP2, \TMP6
1050 pshufd $78, \XMM7, \TMP2
1051 pxor \XMM7, \TMP2
1069 pclmulqdq $0x00, \TMP5, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1078 pxor \TMP2, \TMP6
1084 pshufd $78, \XMM8, \TMP2
1085 pxor \XMM8, \TMP2
1116 pclmulqdq $0x00, \TMP5, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1136 pxor \TMP6, \TMP2
1137 pxor \TMP1, \TMP2
1138 pxor \XMM5, \TMP2
1139 movdqa \TMP2, \TMP3
1141 psrldq $8, \TMP2 # right shift TMP2 2 DWs
1143 pxor \TMP2, \TMP1 # accumulate the results in TMP1:XMM5
1147 movdqa \XMM5, \TMP2
1150 # move XMM5 into TMP2, TMP3, TMP4 in order to perform shifts independently
1151 pslld $31, \TMP2 # packed right shift << 31
1154 pxor \TMP3, \TMP2 # xor the shifted versions
1155 pxor \TMP4, \TMP2
1156 movdqa \TMP2, \TMP5
1158 pslldq $12, \TMP2 # left shift T2 3 DWs
1159 pxor \TMP2, \XMM5
1163 movdqa \XMM5,\TMP2 # make 3 copies of XMM5 into TMP2, TMP3, TMP4
1166 psrld $1, \TMP2 # packed left shift >>1
1169 pxor \TMP3,\TMP2 # xor the shifted versions
1170 pxor \TMP4,\TMP2
1171 pxor \TMP5, \TMP2
1172 pxor \TMP2, \XMM5
1184 .macro GHASH_4_ENCRYPT_4_PARALLEL_dec TMP1 TMP2 TMP3 TMP4 TMP5 \
1231 pshufd $78, \XMM6, \TMP2
1232 pxor \XMM6, \TMP2
1247 pclmulqdq $0x00, \TMP5, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1256 pxor \TMP2, \TMP6
1258 pshufd $78, \XMM7, \TMP2
1259 pxor \XMM7, \TMP2
1277 pclmulqdq $0x00, \TMP5, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1286 pxor \TMP2, \TMP6
1292 pshufd $78, \XMM8, \TMP2
1293 pxor \XMM8, \TMP2
1324 pclmulqdq $0x00, \TMP5, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1348 pxor \TMP6, \TMP2
1349 pxor \TMP1, \TMP2
1350 pxor \XMM5, \TMP2
1351 movdqa \TMP2, \TMP3
1353 psrldq $8, \TMP2 # right shift TMP2 2 DWs
1355 pxor \TMP2, \TMP1 # accumulate the results in TMP1:XMM5
1359 movdqa \XMM5, \TMP2
1362 # move XMM5 into TMP2, TMP3, TMP4 in order to perform shifts independently
1363 pslld $31, \TMP2 # packed right shift << 31
1366 pxor \TMP3, \TMP2 # xor the shifted versions
1367 pxor \TMP4, \TMP2
1368 movdqa \TMP2, \TMP5
1370 pslldq $12, \TMP2 # left shift T2 3 DWs
1371 pxor \TMP2, \XMM5
1375 movdqa \XMM5,\TMP2 # make 3 copies of XMM5 into TMP2, TMP3, TMP4
1378 psrld $1, \TMP2 # packed left shift >>1
1381 pxor \TMP3,\TMP2 # xor the shifted versions
1382 pxor \TMP4,\TMP2
1383 pxor \TMP5, \TMP2
1384 pxor \TMP2, \XMM5
1391 .macro GHASH_LAST_4 TMP1 TMP2 TMP3 TMP4 TMP5 TMP6 \
1397 pshufd $78, \XMM1, \TMP2
1398 pxor \XMM1, \TMP2
1403 pclmulqdq $0x00, \TMP4, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1405 movdqa \TMP2, \XMM1 # result in TMP6, XMMDst, XMM1
1410 pshufd $78, \XMM2, \TMP2
1411 pxor \XMM2, \TMP2
1416 pclmulqdq $0x00, \TMP4, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1419 pxor \TMP2, \XMM1
1425 pshufd $78, \XMM3, \TMP2
1426 pxor \XMM3, \TMP2
1431 pclmulqdq $0x00, \TMP4, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1434 pxor \TMP2, \XMM1 # results accumulated in TMP6, XMMDst, XMM1
1438 pshufd $78, \XMM4, \TMP2
1439 pxor \XMM4, \TMP2
1444 pclmulqdq $0x00, \TMP4, \TMP2 # TMP2 = (a1+a0)*(b1+b0)
1447 pxor \XMM1, \TMP2
1448 pxor \TMP6, \TMP2
1449 pxor \XMMDst, \TMP2
1451 movdqa \TMP2, \TMP4
1453 psrldq $8, \TMP2 # right shift TMP2 2 DWs
1455 pxor \TMP2, \TMP6
1458 movdqa \XMMDst, \TMP2
1461 # move XMMDst into TMP2, TMP3, TMP4 in order to perform 3 shifts independently
1462 pslld $31, \TMP2 # packed right shifting << 31
1465 pxor \TMP3, \TMP2 # xor the shifted versions
1466 pxor \TMP4, \TMP2
1467 movdqa \TMP2, \TMP7
1469 pslldq $12, \TMP2 # left shift TMP2 3 DWs
1470 pxor \TMP2, \XMMDst
1473 movdqa \XMMDst, \TMP2
1477 psrld $1, \TMP2 # packed left shift >> 1
1480 pxor \TMP3, \TMP2 # xor the shifted versions
1481 pxor \TMP4, \TMP2
1482 pxor \TMP7, \TMP2
1483 pxor \TMP2, \XMMDst