bloking while reading from texture memory

I have some different threads from different calculation blocks (or from one). These threads are accessing one adress tex1Dfetch (data, address). Is there any blocking of such reading? Or CUDA does not interfere with the accessing process and the developer can tune all as he want?

If I understand you correctly, there is no problem, each thread can read the same data item
from the texture without problem or need for blocking.