ceph/src/arrow/go/arrow/math/_lib/float64_sse4.s

   1         .text
   2         .intel_syntax noprefix
   3         .file   "_lib/float64.c"
   4         .globl  sum_float64_sse4
   5         .p2align        4, 0x90
   6         .type   sum_float64_sse4,@function
   7 sum_float64_sse4:                       # @sum_float64_sse4
   8 # BB#0:
   9         push    rbp
  10         mov     rbp, rsp
  11         and     rsp, -8
  12         xorpd   xmm0, xmm0
  13         test    rsi, rsi
  14         je      .LBB0_14
  15 # BB#1:
  16         cmp     rsi, 3
  17         jbe     .LBB0_2
  18 # BB#5:
  19         mov     r9, rsi
  20         and     r9, -4
  21         je      .LBB0_2
  22 # BB#6:
  23         lea     r8, [r9 - 4]
  24         mov     eax, r8d
  25         shr     eax, 2
  26         inc     eax
  27         and     rax, 3
  28         je      .LBB0_7
  29 # BB#8:
  30         neg     rax
  31         xorpd   xmm0, xmm0
  32         xor     ecx, ecx
  33         xorpd   xmm1, xmm1
  34         .p2align        4, 0x90
  35 .LBB0_9:                                # =>This Inner Loop Header: Depth=1
  36         movupd  xmm2, xmmword ptr [rdi + 8*rcx]
  37         movupd  xmm3, xmmword ptr [rdi + 8*rcx + 16]
  38         addpd   xmm0, xmm2
  39         addpd   xmm1, xmm3
  40         add     rcx, 4
  41         inc     rax
  42         jne     .LBB0_9
  43         jmp     .LBB0_10
  44 .LBB0_2:
  45         xor     r9d, r9d
  46 .LBB0_3:
  47         lea     rax, [rdi + 8*r9]
  48         sub     rsi, r9
  49         .p2align        4, 0x90
  50 .LBB0_4:                                # =>This Inner Loop Header: Depth=1
  51         addsd   xmm0, qword ptr [rax]
  52         add     rax, 8
  53         dec     rsi
  54         jne     .LBB0_4
  55 .LBB0_14:
  56         movsd   qword ptr [rdx], xmm0
  57         mov     rsp, rbp
  58         pop     rbp
  59         ret
  60 .LBB0_7:
  61         xor     ecx, ecx
  62         xorpd   xmm0, xmm0
  63         xorpd   xmm1, xmm1
  64 .LBB0_10:
  65         cmp     r8, 12
  66         jb      .LBB0_13
  67 # BB#11:
  68         mov     rax, r9
  69         sub     rax, rcx
  70         lea     rcx, [rdi + 8*rcx + 112]
  71         .p2align        4, 0x90
  72 .LBB0_12:                               # =>This Inner Loop Header: Depth=1
  73         movupd  xmm2, xmmword ptr [rcx - 112]
  74         movupd  xmm3, xmmword ptr [rcx - 96]
  75         movupd  xmm4, xmmword ptr [rcx - 80]
  76         movupd  xmm5, xmmword ptr [rcx - 64]
  77         addpd   xmm2, xmm0
  78         addpd   xmm3, xmm1
  79         movupd  xmm6, xmmword ptr [rcx - 48]
  80         movupd  xmm7, xmmword ptr [rcx - 32]
  81         addpd   xmm6, xmm4
  82         addpd   xmm6, xmm2
  83         addpd   xmm7, xmm5
  84         addpd   xmm7, xmm3
  85         movupd  xmm0, xmmword ptr [rcx - 16]
  86         movupd  xmm1, xmmword ptr [rcx]
  87         addpd   xmm0, xmm6
  88         addpd   xmm1, xmm7
  89         sub     rcx, -128
  90         add     rax, -16
  91         jne     .LBB0_12
  92 .LBB0_13:
  93         addpd   xmm0, xmm1
  94         haddpd  xmm0, xmm0
  95         cmp     r9, rsi
  96         jne     .LBB0_3
  97         jmp     .LBB0_14
  98 .Lfunc_end0:
  99         .size   sum_float64_sse4, .Lfunc_end0-sum_float64_sse4
 100
 101
 102         .ident  "Apple LLVM version 9.0.0 (clang-900.0.39.2)"
 103         .section        ".note.GNU-stack","",@progbits