Chroma CUDA
=============

CUDA 101
---------

* http://on-demand.gputechconf.com/gtc/2012/presentations/S0027B-GTC2012-Debugging-MEMCHECK.pdf


* Threads are grouped into blocks
* Blocks are grouped into a grid
* A kernel is executed as a grid of blocks of threads

* `_syncthreads()` enforces synchronization **within a block**


Chroma Shared Geometry
------------------------

Constrains how Chroma uses CUDA perhaps ?