<html>
    <head>
      <base href="http://llvm.org/bugs/" />
    </head>
    <body><table border="1" cellspacing="0" cellpadding="8">
        <tr>
          <th>Bug ID</th>
          <td><a class="bz_bug_link 
          bz_status_NEW "
   title="NEW --- - vector uitofp fails for 64-bit integers"
   href="http://llvm.org/bugs/show_bug.cgi?id=15309">15309</a>
          </td>
        </tr>

        <tr>
          <th>Summary</th>
          <td>vector uitofp fails for 64-bit integers
          </td>
        </tr>

        <tr>
          <th>Product</th>
          <td>new-bugs
          </td>
        </tr>

        <tr>
          <th>Version</th>
          <td>trunk
          </td>
        </tr>

        <tr>
          <th>Hardware</th>
          <td>PC
          </td>
        </tr>

        <tr>
          <th>OS</th>
          <td>Linux
          </td>
        </tr>

        <tr>
          <th>Status</th>
          <td>NEW
          </td>
        </tr>

        <tr>
          <th>Severity</th>
          <td>enhancement
          </td>
        </tr>

        <tr>
          <th>Priority</th>
          <td>P
          </td>
        </tr>

        <tr>
          <th>Component</th>
          <td>new bugs
          </td>
        </tr>

        <tr>
          <th>Assignee</th>
          <td>unassignedbugs@nondot.org
          </td>
        </tr>

        <tr>
          <th>Reporter</th>
          <td>babslachem@gmail.com
          </td>
        </tr>

        <tr>
          <th>CC</th>
          <td>llvmbugs@cs.uiuc.edu
          </td>
        </tr>

        <tr>
          <th>Classification</th>
          <td>Unclassified
          </td>
        </tr></table>
      <p>
        <div>
        <pre>Let's consider following code to convert a 2 x i64 vector into a 2 x float
vector:


; ModuleID = 'opencl-kernel-2N45Jj-2-src-def.ll'
target datalayout =
"e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:32:32-n8:16:32"
target triple = "i386-pc-linux-gnu"

define void @test_convert_float2_ulong2(<2 x i64>* nocapture %src, <2 x float>*
nocapture %dest) noinline {
L.entry:
  %0 = getelementptr <2 x i64>* %src, i32 10
  %1 = load <2 x i64>* %0, align 16
  %2 = uitofp <2 x i64> %1 to <2 x float>
  %3 = getelementptr <2 x float>* %dest, i32 10
  store <2 x float> %2, <2 x float>* %3, align 8
  ret void
}


Using llc as follows:

llc -march=x86 uitofp.ll -o uitofp.s


We got following assembly generated:

    .file    "uitofp.ll"
    .section    .rodata.cst16,"aM",@progbits,16
    .align    16
.LCPI0_0:
    .quad    4841369599423283200     # double 4.503600e+15
    .quad    4841369599423283200     # double 4.503600e+15
    .text
    .globl    test_convert_float2_ulong2
    .align    16, 0x90
    .type    test_convert_float2_ulong2,@function
test_convert_float2_ulong2:             # @test_convert_float2_ulong2
    .cfi_startproc
# BB#0:                                 # %L.entry
    movl    4(%esp), %eax
    movapd    .LCPI0_0, %xmm0
    movapd    160(%eax), %xmm1
    orpd    %xmm0, %xmm1
    subpd    %xmm0, %xmm1
    cvtpd2ps    %xmm1, %xmm0
    movl    8(%esp), %eax
    extractps    $1, %xmm0, 84(%eax)
    movss    %xmm0, 80(%eax)
    ret
.Ltmp0:
    .size    test_convert_float2_ulong2, .Ltmp0-test_convert_float2_ulong2
    .cfi_endproc


    .section    ".note.GNU-stack","",@progbits

If for instance we want to convert vector <0x0010000000000000,
0x0010000000000000> expected result is: <0x59800000, 0x59800000>.
With implemented algorithm we got:

orpd    %xmm0, %xmm1 => xmm1 = <0x4330000000000000, 0x4330000000000000> |
<0x0010000000000000, 0x0010000000000000> = <0x4330000000000000,
0x4330000000000000>
subpd    %xmm0, %xmm1 => xmm1 - xmm0 = <0x0000000000000000, 0x0000000000000000>

Thus final results is <0x00000000, 0x00000000> instead of  <0x59800000,
0x59800000>.</pre>
        </div>
      </p>
      <hr>
      <span>You are receiving this mail because:</span>
      
      <ul>
          <li>You are on the CC list for the bug.</li>
      </ul>
    </body>
</html>