gistfile1.txt

$ cat y.jl
cat: y.jl: No such file or directory
$ cat /tmp/y.jl
function saxpy( a, x, y )
    @simd for i=1:length(x)
        @inbounds y[i] = y[i]+a*x[i]
    end
end

code_native( saxpy, (Float32,Array{Float32,1},Array{Float32,1}) )
$ julia /tmp/y.jl
        .text
Filename: /tmp/y.jl
Source line: 33
        push    RBP
        mov     RBP, RSP
Source line: 33
        mov     RCX, QWORD PTR [RDI + 16]
        xor     EAX, EAX
        test    RCX, RCX
        cmovg   RAX, RCX
        cmp     RAX, 1
        lea     RCX, QWORD PTR [RAX - 1]
        jo      254
        lea     RAX, QWORD PTR [RCX + 1]
        add     RCX, 1
        jo      240
        test    RAX, RAX
        jle     226
Source line: 3
        mov     R8, QWORD PTR [RDI + 8]
        mov     R9, QWORD PTR [RSI + 8]
        xor     EDX, EDX
        mov     R10, RAX
        and     R10, -16
        je      157
        vpshufd XMM1, XMM0, 0           # xmm1 = xmm0[0,0,0,0]
        vinsertf128     YMM1, YMM1, XMM1, 1
        xor     EDI, EDI
        lea     RSI, QWORD PTR [4*RDI]
        mov     RCX, R8
        sub     RCX, RSI
        mov     RDX, R9
        sub     RDX, RSI
                vmovups XMM4, XMMWORD PTR [RDX]
        vmovups XMM5, XMMWORD PTR [RDX + 16]
        vmovups XMM2, XMMWORD PTR [RDX + 32]
        vmovups XMM3, XMMWORD PTR [RDX + 48]
        vinsertf128     YMM4, YMM4, XMM5, 1
        vmovups XMM6, XMMWORD PTR [RCX]
        vmovups XMM7, XMMWORD PTR [RCX + 16]
        vmovups XMM5, XMMWORD PTR [RCX + 32]
        vinsertf128     YMM6, YMM6, XMM7, 1
        vmulps  YMM6, YMM6, YMM1
        add     RDI, -16
        vaddps  YMM4, YMM4, YMM6
        vinsertf128     YMM2, YMM2, XMM3, 1
        vinsertf128     YMM3, YMM5, XMMWORD PTR [RCX + 48], 1
        vmulps  YMM3, YMM3, YMM1
        vaddps  YMM2, YMM2, YMM3
        vextractf128    XMM3, YMM2, 1
        vextractf128    XMM5, YMM4, 1
        mov     RCX, R10
        add     RCX, RDI
        vmovups XMMWORD PTR [RDX + 16], XMM5
        vmovups XMMWORD PTR [RDX], XMM4
        vmovups XMMWORD PTR [RDX + 48], XMM3
        vmovups XMMWORD PTR [RDX + 32], XMM2
        jne     -141
        mov     RDX, R10
        sub     RAX, RDX
        je      37
Source line: 41
        lea     RCX, QWORD PTR [R9 + 4*RDX]
        lea     RDX, QWORD PTR [R8 + 4*RDX]
Source line: 3
        vmulss  XMM1, XMM0, DWORD PTR [RDX]
        vaddss  XMM1, XMM1, DWORD PTR [RCX]
        vmovss  DWORD PTR [RCX], XMM1
Source line: 41
        add     RCX, 4
        add     RDX, 4
        dec     RAX
        jne     -29
Source line: 48
        pop     RBP
        vzeroupper
        ret
Source line: 33
        movabs  RAX, 139833603693952
        mov     RDI, QWORD PTR [RAX]
        movabs  RAX, 139833588617104
        mov     ESI, 33
        call    RAX