Vector's(float4, float8) calculation

I’m just wondering that does vector calculation work for OpenCL
on Linux ?
I tried the calculation on windows xp, and it worked well.
But when i changed to linux, it’s just not working properly.
FIY, I’m using the same GPU for both XP and LINUX.

thx in advance,

It should be the same as the spec is designed for portability. This is likely a problem with the particular implementation you are using.