This adds an extents<N> class which stores an array of n-dimensional
extent values. This can be used to represent an ND-range for launching
an OpenCL kernel or for storing the size/shape of a 2D/3D image.
Also adds a variadic dim() function which provides a concise syntax
for creating extents<N> objects.
refs kylelutz/compute#9
device, context, and queue are initialized statically in `context_setup.hpp`.
With this change all tests are able to complete when an NVIDIA GPU is in
exclusive compute mode.
Side effect of the change:
Time for all tests to complete reduced from 15.71 to 13.03 sec Tesla C2075.
This adds a test for the enqueue_write_buffer_rect() method
in the command_queue class. This method copies a rectangular
region of memory from the host to a device buffer.
This changes the enqueue_nd_range_kernel() method to return an
event object. This allows clients to monitor the progress of a
kernel executing on a device.
This fixes a bug in the event_profiling test case in the
command_queue test. On AMD platforms, the event object
returned from clEnqueueMarker() has no profiling information
associated with it and returns an error code when accessed.
Now, profiling information for a simple write to a device
buffer is checked instead.
