dynamic allocation of shared memory in cudafy

Jan 30, 2014 at 6:35 PM
hello ,
I just wanted to know how shared memory is allocated dynamically while launching a kernel in cudafy.
this is achieved in cuda c like this ..

kernal_name<<<Dg,Db,Ns,S>>>([kernel arguments])

where Ns specifies the number bytes of shared memory that is dynamically allocated per thread block.
Coordinator
Jan 31, 2014 at 8:24 AM
Currently this is not possible and the default shared memory is used.