cuda - Can two blocks be executed in the same warp? -

from cuda c programming guide:

the threads of thread block execute concurrently on 1 multiprocessor, , multiple thread blocks can execute concurrently on 1 multiprocessor. thread blocks terminate, new blocks launched on vacated multiprocessors.

now, have many (e.g. 1000) blocks of few threads each (e.g. 4), , each thread uses few registers multiple threads run together.

i understand if, given warp of e.g. 32 threads, 8 blocks run fill warp, or if not happen.

after reading above quotation, guess may happen if instructions executed same threads in warp regardless blocks in threads (in case, 32 threads 8 blocks happen have same program counter).

but maybe totally wrong. better understand happen if block size smaller warp size.

it not happen. threads in warp come same block.

from programming guide, on assembly of threads warps:

the way block partitioned warps same; each warp contains threads of consecutive, increasing thread ids first warp containing thread 0. thread hierarchy describes how thread ids relate thread indices in block.