Re: [eigen] clang-3.1 issue

[ Thread Index | Date Index | More lists.tuxfamily.org/eigen Archives ]


thanks!

gael

On Wed, Jun 13, 2012 at 9:34 PM, Benjamin Schindler
<bschindler@xxxxxxxxxxx> wrote:
> After experimenting a little bit more, I was convinced and filed a bug:
>
> http://llvm.org/bugs/show_bug.cgi?id=13103
>
> Cheers
> Benjamin
>
>
> On 13.06.2012 20:27, Gael Guennebaud wrote:
>>
>> Indeed, that looks like a clang bug. I've no idea how to workaround it.
>>
>> gael
>>
>> On Wed, Jun 13, 2012 at 1:21 PM, Benjamin Schindler
>> <bschindler@xxxxxxxxxxx>  wrote:
>>>
>>> Hi
>>>
>>> I just stumbled over this issue, may be you have some input on it, but
>>> I'm really thinking I have found a clang bug. Consider the attached code
>>> (I used latest tip of Eigen). It fails to compile with
>>>
>>> test.cpp:11:18: error: use of overloaded operator '=' is ambiguous (with
>>> operand types 'typename FixedSegmentReturnType<3>::Type' (aka
>>> 'VectorBlock<Eigen::Matrix<float, 4, 1, 0, 4, 1>, 3>') and 'const
>>> ConstantReturnType' (aka 'const
>>>      CwiseNullaryOp<internal::scalar_constant_op<Scalar>,
>>> Eigen::Matrix<float, 3, 1, 0, 3, 1>  >'))
>>>        input.head<3>() = Vector3f::Zero();
>>>        ~~~~~~~~~~~~~~~ ^ ~~~~~~~~~~~~~~~~
>>> test.cpp:17:2: note: in instantiation of function template
>>> specialization 'myFunction<float>' requested here
>>>        myFunction<float>();
>>>        ^
>>> /local/bschindler/projects/eigen/Eigen/src/Core/MatrixBase.h:169:14:
>>> note: candidate function [with OtherDerived =
>>> Eigen::CwiseNullaryOp<Eigen::internal::scalar_constant_op<float>,
>>> Eigen::Matrix<float, 3, 1, 0, 3, 1>  >]
>>>    Derived&  operator=(const DenseBase<OtherDerived>&  other);
>>>             ^
>>> /local/bschindler/projects/eigen/Eigen/src/Core/MatrixBase.h:172:14:
>>> note: candidate function [with OtherDerived =
>>> Eigen::CwiseNullaryOp<Eigen::internal::scalar_constant_op<float>,
>>> Eigen::Matrix<float, 3, 1, 0, 3, 1>  >]
>>>    Derived&  operator=(const EigenBase<OtherDerived>&  other);
>>>
>>>
>>> Now, If I move lines 10 and 11 into the main function it compiles.
>>> clang bug?
>>>
>>> Cheers
>>> Benjamin
>>
>>
>



Mail converted by MHonArc 2.6.19+ http://listengine.tuxfamily.org/