so for example if there was an idle GPU and Q3 take it and then there is a task comes to Q2 which we specified 3GPU but now the Q3 is taken some of these GPU what will happen

This is a standard "race" the first one to come will "grab" the GPU and the other will wait for it.
I'm pretty sure enterprise edition has preemption support, but this is not currently part of the open source version (btw: also the dynamic GPU allocation, I think, is part of the enterprise tier, in the opensource version the allocation is static, so in our case it would be static 3,3,2)