[LLVMbugs] [Bug 3383] New: Assertion failed: Only trivial vector-to-scalar conversions should get here!

[bugzilla-daemon at cs.uiuc.edu]

http://llvm.org/bugs/show_bug.cgi?id=3383

           Summary: Assertion failed: Only trivial vector-to-scalar
                    conversions should get here!
           Product: new-bugs
           Version: unspecified
          Platform: PC
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: new bugs
        AssignedTo: unassignedbugs at nondot.org
        ReportedBy: lennart at augustsson.net
                CC: llvmbugs at cs.uiuc.edu


With LLVM 2.4, generating assembly (x86) for this generates a failed assertion:

define <4 x i1> @fun(<4 x float>) {
_L1:
        %1 = extractelement <4 x float> %0, i32 0
        %2 = fcmp olt float %1, 0.000000e+00
        %3 = insertelement <4 x i1> undef, i1 %2, i32 0
        %4 = extractelement <4 x float> %0, i32 1
        %5 = fcmp olt float %4, 0.000000e+00
        %6 = insertelement <4 x i1> %3, i1 %5, i32 1
        %7 = extractelement <4 x float> %0, i32 2
        %8 = fcmp olt float %7, 0.000000e+00
        %9 = insertelement <4 x i1> %6, i1 %8, i32 2
        %10 = extractelement <4 x float> %0, i32 3
        %11 = fcmp olt float %10, 0.000000e+00
        %12 = insertelement <4 x i1> %9, i1 %11, i32 3
        ret <4 x i1> %12
}


-- 
Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.