[LLVMbugs] [Bug 16201] New: Code gen for contracted FP operation wrong

[bugzilla-daemon at llvm.org]

http://llvm.org/bugs/show_bug.cgi?id=16201

            Bug ID: 16201
           Summary: Code gen for contracted FP operation wrong
           Product: clang
           Version: unspecified
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: -New Bugs
          Assignee: unassignedclangbugs at nondot.org
          Reporter: tydeman at tybor.com
                CC: llvmbugs at cs.uiuc.edu
    Classification: Unclassified

Code generation appears to be wrong for the contracted floating-point
operation:
  p = x*y;
  m = -p;
  z = -m - (x*y);
where all the variables are of type float or double.  The other three forms:
  z = x*y - p;
  z = x*y + m;
  z = p - (x*y);
appear to be OK.

This is clang 3.3-0.4 on Linux Fedora Core 19 in 32-bit mode on an Intel
Core 2 Duo CPU.

Command line options:  -std=c1x -O0 -pedantic

-- 
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/20130603/c0047320/attachment.html>