Hi,
first of all sorry for my bad english.
I’ve just started learning about CUDA and i read (on this page: Programming Guide :: CUDA Toolkit Documentation) information that
“Maximum number of instructions per kernel” is 2 or 512 mln.
Could anyone tell me what does it mean?
For example if i have over 512 mln data on GPU and i want to each thread do something on all of this data (For example i want to count: sum += data[i]*threadIdx.x;) there will be error, because there will be over 512 mln instructions?