[LLVMbugs] [Bug 21138] New vector shuffle lowering produces an insertf128 + blend instead of a single insertf128

[bugzilla-daemon at llvm.org]

http://llvm.org/bugs/show_bug.cgi?id=21138

Quentin Colombet <qcolombet at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|FIXED                       |---

--- Comment #2 from Quentin Colombet <qcolombet at apple.com> ---
Don't know if I missed something, but r219133 still has the issue.
Though, this is a bit different, i.e., instead of blendps we have blendpd:
     vextractf128    $1, %ymm1, %xmm2
     vshufps    $-35, %xmm2, %xmm1, %xmm1 ## xmm1 = xmm1[1,3],xmm2[1,3]
+    vinsertf128    $1, %xmm1, %ymm0, %ymm1
     vextractf128    $1, %ymm0, %xmm2
     vshufps    $-35, %xmm2, %xmm0, %xmm0 ## xmm0 = xmm0[1,3],xmm2[1,3]
-    vinsertf128    $1, %xmm1, %ymm0, %ymm0
+    vblendpd    $12, %ymm1, %ymm0, %ymm0 ## ymm0 = ymm0[0,1],ymm1[2,3]
     retq

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20141006/c5639832/attachment.html>