Cast indices to intp rather than intc.

The intent was correctly described in the comment - to cast to int32 on 32-bit machines and to int64
on 64-bit.  However, numpy.intc is not correct type to achieve that because it has the size of C int and
is 32-bit always.  The correct type is numpy.intp which is defined as "large enough to fit a pointer."

See http://docs.scipy.org/doc/numpy/reference/arrays.scalars.html .

Make compile_str able to compile a lib that is not a Python module

Added a verbosity flag to device_malloc.

Fix cachelock

Set the file in the compiledir, not the case. Some user set the compiledir directly, not the base_compiledir and this make that sometimes, the base_compiledir isn't working.

It is a todo to find a better fix, but to don't cause probleme, here is a work around.

Shorter error

Minor fixes

Make sure preallocated output is contiguous before using it

Take the compiledir lock only if the op generate c code.

Make check_preallocated_output work on fct outputs

Also set Python Exceptions before failing.

grad method for GpuContiguous

fix a NotScalarConstantError issue

Add test for when b is not constant

To be able to use negative offsets in indexing, threadIdx.x (and y)
should be cast to signed int.

Issue #1193: Fixed theano-nose --batch.

It may be slower than allocated new memory and calling memset,
but if that non-contiguous memory was provided, it is probable
that the output will end up being copied into that non-contiguous
memory, and that would be even slower.