Tracing the GPU usage

Hello,

Can give me some tips to trace the GPU usage on a solaris8 update7 box?
I try to understand why a Cadcam app is so slow on my Solaris box compared to a win$ box. I guess it comes from the poor graphic card I have but i would like to emphasize it.

My bench is 150sec long on a Solaris while 33sec long on Win$
Using vmstat and iostat I can see that there is no disk swap that could slow the app. The cpu is spent, at 95%, as user CPU. The truss -c cmd shows that the overall system time is 13sec (over 150sec) with 12 secs of ioctl!
As it's a graphic app I guess the system is wating for "GPU results". HOW can I trace it from a system point of view (I don't want to use development mathematics library to trace the process)? The only stuff I have found is fbconfig -prconf that gaves me the video memory usage... :o

Many thanks for your help.