<html>
    <head>
      <base href="https://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 --- - [x86] bad codegen while extracting vector elements"
   href="https://llvm.org/bugs/show_bug.cgi?id=25543">25543</a>
          </td>
        </tr>

        <tr>
          <th>Summary</th>
          <td>[x86] bad codegen while extracting vector elements
          </td>
        </tr>

        <tr>
          <th>Product</th>
          <td>libraries
          </td>
        </tr>

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

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

        <tr>
          <th>OS</th>
          <td>All
          </td>
        </tr>

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

        <tr>
          <th>Severity</th>
          <td>normal
          </td>
        </tr>

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

        <tr>
          <th>Component</th>
          <td>Backend: X86
          </td>
        </tr>

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

        <tr>
          <th>Reporter</th>
          <td>spatel+llvm@rotateright.com
          </td>
        </tr>

        <tr>
          <th>CC</th>
          <td>llvm-bugs@lists.llvm.org
          </td>
        </tr>

        <tr>
          <th>Classification</th>
          <td>Unclassified
          </td>
        </tr></table>
      <p>
        <div>
        <pre>It may be possible to further reduce this, but here's a small C example that
shows the problem(s):

$ cat shifty.c
#include <immintrin.h>

void foo(__m128i v, int bits, short *s0, short *s1) {
  int i[4];
  _mm_storeu_si128((__m128i *)i, v); // get vector bits as ints
  if (bits & 1) { *s0 = i[0]; }      // conditionally store part of elt 0
  if (bits & 2) { *s1 = i[1]; }      // conditionally store part of elt 1
}

Testing with:
$ ./clang -v
clang version 3.8.0 (trunk 253043)
Target: x86_64-apple-darwin15.0.0


$ ./clang -fomit-frame-pointer -O2 shifty.c -S -o -
_foo:
    movd    %xmm0, %rax   # Wrong disasm? That's a 64-bit move (movq).
    testb    $1, %dil
    je    LBB0_2
    movw    %ax, (%rsi)
LBB0_2:
    testb    $2, %dil
    je    LBB0_4

# Things get weird...why are we messing with elements 2 and 3?

    pshufd    $78, %xmm0, %xmm0       ## xmm0 = xmm0[2,3,0,1]
    movd    %xmm0, %rcx
    shldq    $32, %rax, %rcx   # Could have just shifted %rax left?
    movw    %cx, (%rdx)
LBB0_4:
    retq</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>