def test_rmm_csv_log(dtype, nelem):
# data
h_in = np.full(nelem, 3.2, dtype)
d_in = rmm.to_device(h_in)
d_result = rmm.device_array_like(d_in)
d_result.copy_to_device(d_in)
csv = rmm.csv_log()
assert (csv.find("Event Type,Device ID,Address,Stream,Size (bytes),"
"Free Memory,Total Memory,Current Allocs,Start,End,"
"Elapsed,Location") >= 0)
# Very simple allocation / free test # # Expected output is along the lines of: # # $ python simple01.py # Event Type,Device ID,Address,Stream,Size (bytes),Free Memory,Total Memory,Current Allocs,Start,End,Elapsed,Location # Alloc,0,0x7fae06600000,0,80,0,0,1,1.10549,1.1074,0.00191666,/home/nfs/gmarkall/numbadev/numba/numba/cuda/cudadrv/driver.py:683 # Free,0,0x7fae06600000,0,0,0,0,0,1.10798,1.10921,0.00122238,/home/nfs/gmarkall/numbadev/numba/numba/utils.py:678 # # One allocation results from the creation of d_a in cuda.to_device. # On free results from the deletion of d_a. import rmm import numpy as np from numba import cuda rmm.use_rmm_for_numba() a = np.zeros(10) d_a = cuda.to_device(a) del (d_a) print(rmm.csv_log())