Lines Matching refs:aptr
1086 my $aptr="%rsi"; # const BN_ULONG *aptr,
1146 sub %r11,%rbp # align with $aptr
1210 mov $aptr,$rptr
1319 lea ($aptr,$num),$aptr # end of a[] buffer, ($aptr,$i)=&ap[2]
1324 mov -32($aptr,$i),$a0 # a[0]
1326 mov -24($aptr,$i),%rax # a[1]
1328 mov -16($aptr,$i),$ai # a[2]
1345 mov -8($aptr,$i),$ai # a[3]
1364 mov ($aptr,$j),$ai # a[4]
1374 mov 8($aptr,$j),$ai # a[5]
1391 mov 16($aptr,$j),$ai # a[6]
1407 mov 24($aptr,$j),$ai # a[7]
1448 mov -32($aptr,$i),$a0 # a[0]
1450 mov -24($aptr,$i),%rax # a[1]
1452 mov -16($aptr,$i),$ai # a[2]
1474 mov -8($aptr,$i),$ai # a[3]
1497 mov ($aptr,$j),$ai # a[4]
1510 mov 8($aptr,$j),$ai # a[5]
1553 mov -32($aptr),$a0 # a[0]
1555 mov -24($aptr),%rax # a[1]
1557 mov -16($aptr),$ai # a[2]
1573 mov -8($aptr),$ai # a[3]
1594 mov -16($aptr),%rax # a[2]
1620 mov -16($aptr,$i),%rax # a[0]
1636 mov -8($aptr,$i),%rax # a[i+1] # prefetch
1653 mov 0($aptr,$i),%rax # a[i+1] # prefetch
1675 mov -8($aptr,$i),%rax # a[i+1] # prefetch
1692 mov 0($aptr,$i),%rax # a[i+1] # prefetch
1709 mov 8($aptr,$i),%rax # a[i+1] # prefetch
1726 mov 16($aptr,$i),%rax # a[i+1] # prefetch
1747 mov -8($aptr),%rax # a[i+1] # prefetch
2055 movq %xmm1,$aptr # prepare for back-to-back call
2157 sub %r11,%rbp # align with $aptr
2210 movdqu ($aptr),%xmm1
2211 movdqu 16($aptr),%xmm2
2212 movdqu 32($aptr),%xmm3
2214 movdqu 48($aptr),%xmm4
2216 .byte 0x48,0x8d,0xb6,0x40,0x00,0x00,0x00 # lea 64($aptr),$aptr
2338 sub %r11,%rbp # align with $aptr
2426 my ($aptr, $bptr, $nptr, $tptr, $mi, $bi, $zero, $num)=
2520 mulx 0*8($aptr),$mi,%rax # a[0]*b[0]
2521 mulx 1*8($aptr),%r11,%r12 # a[1]*b[0]
2523 mulx 2*8($aptr),%rax,%r13 # ...
2526 mulx 3*8($aptr),%rax,%r14
2535 lea 4*8($aptr),$aptr
2562 mulx 0*8($aptr),%r10,%rax # a[4]*b[0]
2564 mulx 1*8($aptr),%r11,%r14 # a[5]*b[0]
2566 mulx 2*8($aptr),%r12,%rax # ...
2568 mulx 3*8($aptr),%r13,%r14
2573 lea 4*8($aptr),$aptr
2601 lea ($aptr,$num),$aptr # rewind $aptr
2640 mulx 0*8($aptr),$mi,%r11 # a[0]*b[i]
2643 mulx 1*8($aptr),%r14,%r12 # a[1]*b[i]
2646 mulx 2*8($aptr),%r15,%r13 # ...
2649 mulx 3*8($aptr),%rdx,%r14
2653 lea 4*8($aptr),$aptr
2687 mulx 0*8($aptr),%r10,%rax # a[4]*b[i]
2690 mulx 1*8($aptr),%r11,%r14 # a[5]*b[i]
2693 mulx 2*8($aptr),%r12,%rax # ...
2696 mulx 3*8($aptr),%r13,%r14
2702 lea 4*8($aptr),$aptr
2735 lea ($aptr,$num),$aptr # rewind $aptr
2769 my $aptr="%rsi"; # const BN_ULONG *aptr,
2821 sub %r11,%rbp # align with $aptr
2887 mov $aptr,$rptr
2970 lea ($aptr,$num),$aaptr
2972 mov $aaptr,8+8(%rsp) # save end of $aptr
2992 mov 0*8($aptr),%rdx # a[0], modulo-scheduled
3006 mulx 1*8($aptr),%r8,%rax # a[1]*a[0]
3009 mulx 2*8($aptr),%r9,%rax # a[2]*a[0]
3012 .byte 0xc4,0xe2,0xab,0xf6,0x86,0x18,0x00,0x00,0x00 # mulx 3*8($aptr),%r10,%rax # ...
3015 .byte 0xc4,0xe2,0xa3,0xf6,0x86,0x20,0x00,0x00,0x00 # mulx 4*8($aptr),%r11,%rax
3018 mulx 5*8($aptr),%r12,%rax
3021 mulx 6*8($aptr),%r13,%rax
3024 mulx 7*8($aptr),%r14,%r15
3025 mov 1*8($aptr),%rdx # a[1]
3035 mulx 2*8($aptr),%r8,%rbx # a[2]*a[1]
3036 mulx 3*8($aptr),%r9,%rax # a[3]*a[1]
3039 mulx 4*8($aptr),%r10,%rbx # ...
3042 .byte 0xc4,0xe2,0xa3,0xf6,0x86,0x28,0x00,0x00,0x00 # mulx 5*8($aptr),%r11,%rax
3045 .byte 0xc4,0xe2,0x9b,0xf6,0x9e,0x30,0x00,0x00,0x00 # mulx 6*8($aptr),%r12,%rbx
3048 .byte 0xc4,0x62,0x93,0xf6,0xb6,0x38,0x00,0x00,0x00 # mulx 7*8($aptr),%r13,%r14
3049 mov 2*8($aptr),%rdx # a[2]
3059 mulx 3*8($aptr),%r8,%rbx # a[3]*a[2]
3060 mulx 4*8($aptr),%r9,%rax # a[4]*a[2]
3063 mulx 5*8($aptr),%r10,%rbx # ...
3066 .byte 0xc4,0xe2,0xa3,0xf6,0x86,0x30,0x00,0x00,0x00 # mulx 6*8($aptr),%r11,%rax
3069 .byte 0xc4,0x62,0x9b,0xf6,0xae,0x38,0x00,0x00,0x00 # mulx 7*8($aptr),%r12,%r13
3071 mov 3*8($aptr),%rdx # a[3]
3077 mulx 4*8($aptr),%r8,%rax # a[4]*a[3]
3081 mulx 5*8($aptr),%r9,%rbx # a[5]*a[3]
3084 mulx 6*8($aptr),%r10,%rax # ...
3087 mulx 7*8($aptr),%r11,%r12
3088 mov 4*8($aptr),%rdx # a[4]
3089 mov 5*8($aptr),%r14 # a[5]
3092 mov 6*8($aptr),%r15 # a[6]
3101 mov 7*8($aptr),%r8 # a[7]
3116 lea 8*8($aptr),$aptr
3127 cmp 8+8(%rsp),$aptr
3141 lea ($aptr),$aaptr
3145 mov -64($aptr),%rdx # a[0]
3187 mov 8($aptr,%rcx,8),%rdx # a[i]
3203 mov -64($aptr),%rdx
3226 mov 0*8($aptr),%rdx # a[8], modulo-scheduled
3268 mov ($aptr,$i),%rdx # a[0]
3283 .byte 0x48,0x8b,0x94,0x0e,0x08,0x00,0x00,0x00 # mov 8($aptr,$i),%rdx # a[i+1] # prefetch
3294 mov 16($aptr,$i),%rdx # a[i+2] # prefetch
3305 mov 24($aptr,$i),%rdx # a[i+3] # prefetch
3318 .byte 0x48,0x8b,0x94,0x0e,0x00,0x00,0x00,0x00 # mov 0($aptr,$i),%rdx # a[i+4] # prefetch
3576 movq %xmm1,$aptr # prepare for back-to-back call