I decided on OpenCL purely for portability reasons. If I had written a Gallium state tracker specifically for WebM video, the state tracker would only really be useful with Mesa. By doing OpenCL, I don't have to write OpenGL/GLSL (I have basic familiarity with it, but I'm no expert), and I also get the ability to support Mac/Linux/Windows on any usable hardware. As a bonus, Mesa will be able to handle WebM over OpenCL as soon as the Clover project gets to a usable state.
Veerappan , you might want to pop over to mesa-dev and comment on this to push your Clover wish i assume you do know that code base and can think of some generic 30 minute prototypes he might try out and add before thinking about speed and final code etc
the mesa-dev dev mentor's as relates to new GSOC code seem very reluctant to speak more than a few word's, to much tunnel vision for 2D/3D computer generated code perhaps and no real interest in core video playback OC.
and giving some practical examples of what to consider including and how one might think about extending the current frame work well done, a real mentor helping to prod push and inspire contribution.