def test_run_cuda(lt_ctx, mask_cupy):
# The cupy module is set to None in mask_cupy fixture so that
# any use of it will raise an error
with pytest.raises(ModuleNotFoundError):
import cupy # NOQA: F401
data = _mk_random(size=(16, 16, 16, 16))
ds = lt_ctx.load("memory", data=data)
use_cpu = bae.get_use_cpu()
use_cuda = bae.get_use_cuda()
backend = bae.get_device_class()
with mock.patch.dict(os.environ, {'LIBERTEM_USE_CUDA': "23"}):
# This should set the same environment variable as the mock above
# so that it will be unset after the "with"
bae.set_use_cuda(23)
res = lt_ctx.run_udf(
udf=DebugDeviceUDF(backends=('cuda', 'numpy')),
dataset=ds
)
for val in res['device_id'].data[0].values():
print(val)
assert val['cpu'] is None
assert val['cuda'] == 23
# We make sure that the mocking was successful, i.e.
# restored the previous state
assert use_cpu == bae.get_use_cpu()
assert use_cuda == bae.get_use_cuda()
assert backend == bae.get_device_class()
assert np.all(res['device_class'].data == 'cuda')
assert np.allclose(res['on_device'].data, data.sum(axis=(0, 1)))
def test_run_cupy(lt_ctx, mock_cupy):
data = _mk_random(size=(16, 16, 16, 16))
ds = lt_ctx.load("memory", data=data)
use_cpu = bae.get_use_cpu()
use_cuda = bae.get_use_cuda()
backend = bae.get_device_class()
with mock.patch.dict(os.environ, {'LIBERTEM_USE_CUDA': "23"}):
# This should set the same environment variable as the mock above
# so that it will be unset after the "with"
bae.set_use_cuda(23)
# add `numpy.cuda` so we can make `numpy` work as a mock replacement for `cupy`
with mock.patch('numpy.cuda', return_value=MockCuda, create=True):
res = lt_ctx.run_udf(
udf=DebugDeviceUDF(backends=('cupy', 'numpy')),
dataset=ds
)
for val in res['device_id'].data[0].values():
assert val['cpu'] is None
assert val['cuda'] == 23
# We make sure that the mocking was successful, i.e.
# restored the previous state
assert use_cpu == bae.get_use_cpu()
assert use_cuda == bae.get_use_cuda()
assert backend == bae.get_device_class()
assert np.all(res['device_class'].data == 'cuda')
assert np.allclose(res['on_device'].data, data.sum(axis=(0, 1)))
def test_run_numpy(lt_ctx, mask_cupy):
data = _mk_random(size=(16, 16, 16, 16))
ds = lt_ctx.load("memory", data=data)
use_cpu = bae.get_use_cpu()
use_cuda = bae.get_use_cuda()
backend = bae.get_device_class()
with mock.patch.dict(os.environ, {'LIBERTEM_USE_CPU': "42"}):
# This should set the same environment variable as the mock above
# so that it will be unset after the "with"
bae.set_use_cpu(42)
res = lt_ctx.run_udf(udf=DebugDeviceUDF(), dataset=ds)
for val in res['device_id'].data[0].values():
assert val['cpu'] == 42
assert val['cuda'] is None
# We make sure that the mocking was successful, i.e.
# restored the previous state
assert use_cpu == bae.get_use_cpu()
assert use_cuda == bae.get_use_cuda()
assert backend == bae.get_device_class()
assert np.all(res['device_class'].data == 'cpu')
assert np.allclose(res['on_device'].data, data.sum(axis=(0, 1)))
def run_for_partition(self, partition: Partition, roi, corrections):
with set_num_threads(1):
try:
previous_id = None
device_class = get_device_class()
# numpy_udfs and cupy_udfs contain references to the objects in
# self._udfs
numpy_udfs, cupy_udfs = self._udf_lists(device_class)
# Will only be populated if actually on CUDA worker
# and any UDF supports 'cupy' (and not 'cuda')
if cupy_udfs:
# Avoid importing if not used
import cupy
device = get_use_cuda()
previous_id = cupy.cuda.Device().id
cupy.cuda.Device(device).use()
(meta, tiling_scheme,
dtype) = self._init_udfs(numpy_udfs, cupy_udfs, partition,
roi, corrections, device_class)
# print("UDF TilingScheme: %r" % tiling_scheme.shape)
partition.set_corrections(corrections)
self._run_udfs(numpy_udfs, cupy_udfs, partition, tiling_scheme,
roi, dtype)
self._wrapup_udfs(numpy_udfs, cupy_udfs, partition)
finally:
if previous_id is not None:
cupy.cuda.Device(previous_id).use()
# Make sure results are in the same order as the UDFs
return tuple(udf.results for udf in self._udfs)
def get_available_workers(self):
resources = {"compute": 1, "CPU": 1}
if get_use_cuda() is not None:
resources["CUDA"] = 1
return WorkerSet(
[Worker(name='inline', host='localhost', resources=resources)])
def process_partition(self, partition):
cpu = bae.get_use_cpu()
cuda = bae.get_use_cuda()
self.results.device_id[0][self.meta.slice] = {"cpu": cpu, "cuda": cuda}
self.results.on_device[:] += self.xp.sum(partition, axis=0)
self.results.device_class[:] = self.meta.device_class
self.results.backend[0][self.meta.slice] = str(self.xp)
print(f"meta device_class {self.meta.device_class}")
def get_available_workers(self):
resources = {"compute": 1, "CPU": 1}
if get_use_cuda() is not None:
resources["CUDA"] = 1
return WorkerSet([
Worker(name='concurrent',
host='localhost',
resources=resources,
nthreads=1)
])
else:
devices = detect()
return WorkerSet([
Worker(
name='concurrent',
host='localhost',
resources=resources,
nthreads=len(devices['cpus']),
)
])
