[llvm-bugs] [Bug 34992] New: Fatal error: Offset not zero at the point of scalar access

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=34992

            Bug ID: 34992
           Summary: Fatal error: Offset not zero at the point of scalar
                    access
           Product: new-bugs
           Version: trunk
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: new bugs
          Assignee: unassignedbugs at nondot.org
          Reporter: douglas_yung at playstation.sony.com
                CC: llvm-bugs at lists.llvm.org

Following r315984, one of our internal tests started to fail with the following
error from the compiler:

Offset not zero at the point of scalar access
  %2 = load float, float* %india, align 4, !tbaa !6
!6 = !{!7, !9, i64 4}
4
fatal error: error in backend: Broken function found, compilation aborted!

You can reproduce this by compiling the following code with optimization’s
enabled:

Clang -cc1 -emit-obj -O2 reduced.cpp

Where reduced.cpp is the following code:

/* reduced.cpp */
class alpha;
namespace bravo {
  namespace charlie { 
    void delta(alpha* element, char* name, float var); 
  }

  namespace foxtrot {
    class gulf { public: float hotel; float india; };
    class juliet { public: gulf kilo[4]; };
  }

  namespace charlie {
    class lima { public: bravo::foxtrot::juliet mike; };
    void november( alpha* oscar, lima * papa ) {
      alpha* quebec ;
      delta(quebec, "romeo", papa->mike.kilo->india);
    }
  }
}

Thanks to Sunil for helping reduce this test for me.

-- 
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/20171018/335aad7b/attachment-0001.html>