[LLVMbugs] [Bug 17496] New: __builtin_copysignf fails to produce result -0.0f

[bugzilla-daemon at llvm.org]

http://llvm.org/bugs/show_bug.cgi?id=17496

            Bug ID: 17496
           Summary: __builtin_copysignf fails to produce result -0.0f
           Product: new-bugs
           Version: unspecified
          Hardware: All
                OS: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
          Assignee: unassignedbugs at nondot.org
          Reporter: nrink at broadcom.com
                CC: llvmbugs at cs.uiuc.edu
    Classification: Unclassified

I am using the X86 backend, but have "Expand" set as the lowering operation for
the following ISD node types:

FNEG,
FABS,
FCOPYSIGN

(in X86ISelLowering.cpp). With these node expansions, the following program
outputs "0x00000000":

#include <stdio.h>

float x =  0.0f;
float y = -2.0f;

int main()
{
  float r = __builtin_copysignf(x, y);
  printf("0x%08x\n", *(unsigned int*)&r);
  return 0;
}

When compiled with gcc, it gives "0x80000000", as I would expect.

-- 
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/20131007/ee1c2857/attachment.html>