Removed use of enable_if for matrix operations. Now everything compiles cleanly again on Visual Studio 2005.

After issues with Visual Studio 2012 were resolved, Visual Studio 2005 now has some problems with enable_if<>.
Hence, a couple of enable_if were replaced by overloads or removed completely.

Tests with Visual Studio 2012 to come...

Other tutorials and tests still to be revived.

Matrix operations still broken. Work in progress.

Allows to support vector, vector_range, and vector_slice with less code.
Currently only examples/tutorial/blas1.{cpp,cu} compiles. Work in progress.

Reverted 128x128 work layout for CSR SpMV. Instead, increased search range of autotuner (better effect)

Thread config 128x128 for CSR matrix-vector product is now applied right there. This gives the best of both worlds.

Old settings based on core-count were based on ancient OpenCL SDKs and are no longer appropriate.

* Added guard for not using AMD GEMM kernels if device has less than 20 kB of local memory.
* Fixed a warning for GEMM kernels (fast NVIDIA version).

Added fast GEMM kernels for AMD Tahiti based on input from Philippe's autotuner. Only works for square matrices with dimensions being a multiple of 256.

Fixed a warning (/* in comment) in CL/cl_gl_ext.h (staying with OpenCL 1.1. The fix is already in OpenCL 1.2 headers)

Replaced all uses of size_t by std::size_t (exception: viennacl/generator/*, which will be replaced soon anyway)

* Fixed some corner cases for BLAS-1-type operations on vectors.
* Removed bottleneck in sparse-test-XYZ (manually transposing ublas-matrix).

* Fixed an access violation in copy() from STL to device for compressed_matrix with empty rows.

Better OpenMP support for vector operations. Not using OpenMP below 5000 entries to avoid unnecessary startup-overhead.

Added support for Xeon Phi (using 128x128 work items). Adjusted number of work groups on CPU to power of two (six-core CPUs otherwise lead to problems with reductions).