Lines Matching refs:ap
83 my $ap="%rsi"; # const BN_ULONG *ap,
173 sub \$-128, $ap
212 vmovdqu 32*1-128($ap), $ACC1
213 vmovdqu 32*2-128($ap), $ACC2
214 vmovdqu 32*3-128($ap), $ACC3
215 vmovdqu 32*4-128($ap), $ACC4
216 vmovdqu 32*5-128($ap), $ACC5
217 vmovdqu 32*6-128($ap), $ACC6
218 vmovdqu 32*7-128($ap), $ACC7
219 vmovdqu 32*8-128($ap), $ACC8
234 vpbroadcastq 32*0-128($ap), $B1
252 vpmuludq 32*0-128($ap), $B1, $ACC0
253 vpbroadcastq 32*1-128($ap), $B2
270 vpbroadcastq 32*2-128($ap), $B1
273 mov $ap, $tpa
283 vpmuludq 32*0-128($ap), $B1, $ACC0
306 vpmuludq 32*1-128($ap), $B2, $TEMP0
327 vpmuludq 32*2-128($ap), $B1, $TEMP2
346 vpmuludq 32*3-128($ap), $B2, $TEMP0
363 vpmuludq 32*4-128($ap), $B1, $TEMP0
379 vpmuludq 32*5-128($ap), $B2, $TEMP2
392 vpmuludq 32*6-128($ap), $B1, $TEMP0
405 vpmuludq 32*7-128($ap), $B2, $TEMP0
410 vpmuludq 32*8-128($ap), $ACC0, $ACC7
795 mov $rp, $ap
840 my $ap="%rsi"; # const BN_ULONG *ap,
925 # cause severe performance degradation here, so if $ap does
927 # is advised to lay down $ap and $bp next to each other, so
930 mov $ap, $tmp
934 mov $ap, $tmp
935 cmovnz $bp, $ap
939 sub \$-128,$ap # size optimization
1005 imulq -128($ap), %rax
1008 imulq 8-128($ap), $r1
1016 imulq 16-128($ap), $r2
1020 imulq 24-128($ap), $r3
1022 vpmuludq 32*1-128($ap),$Bi,$TEMP0
1025 vpmuludq 32*2-128($ap),$Bi,$TEMP1
1028 vpmuludq 32*3-128($ap),$Bi,$TEMP2
1031 vpmuludq 32*4-128($ap),$Bi,$TEMP0
1033 vpmuludq 32*5-128($ap),$Bi,$TEMP1
1035 vpmuludq 32*6-128($ap),$Bi,$TEMP2
1037 vpmuludq 32*7-128($ap),$Bi,$TEMP0
1040 vpmuludq 32*8-128($ap),$Bi,$TEMP1
1079 imulq -128($ap),%rax
1081 vmovdqu -8+32*1-128($ap),$TEMP1
1083 imulq 8-128($ap),%rax
1085 vmovdqu -8+32*2-128($ap),$TEMP2
1093 imulq 16-128($ap),%rbx
1097 vmovdqu -8+32*3-128($ap),$TEMP0
1101 vmovdqu -8+32*4-128($ap),$TEMP1
1104 vmovdqu -8+32*5-128($ap),$TEMP2
1107 vmovdqu -8+32*6-128($ap),$TEMP0
1110 vmovdqu -8+32*7-128($ap),$TEMP1
1113 vmovdqu -8+32*8-128($ap),$TEMP2
1116 vmovdqu -8+32*9-128($ap),$ACC9
1163 vmovdqu -16+32*1-128($ap),$TEMP0
1165 imulq -128($ap),%rax
1168 vmovdqu -16+32*2-128($ap),$TEMP1
1173 imulq 8-128($ap),%rbx
1177 vmovdqu -16+32*3-128($ap),$TEMP2
1181 vmovdqu -16+32*4-128($ap),$TEMP0
1184 vmovdqu -16+32*5-128($ap),$TEMP1
1187 vmovdqu -16+32*6-128($ap),$TEMP2
1190 vmovdqu -16+32*7-128($ap),$TEMP0
1193 vmovdqu -16+32*8-128($ap),$TEMP1
1196 vmovdqu -16+32*9-128($ap),$TEMP2
1236 vmovdqu -24+32*1-128($ap),$TEMP0
1239 vmovdqu -24+32*2-128($ap),$TEMP1
1243 imulq -128($ap),%rbx
1252 vmovdqu -24+32*3-128($ap),$TEMP2
1256 vmovdqu -24+32*4-128($ap),$TEMP0
1259 vmovdqu -24+32*5-128($ap),$TEMP1
1262 vmovdqu -24+32*6-128($ap),$TEMP2
1265 vmovdqu -24+32*7-128($ap),$TEMP0
1268 vmovdqu -24+32*8-128($ap),$TEMP1
1271 vmovdqu -24+32*9-128($ap),$TEMP2
