1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
|
# file scalene/scalene_nvidia_gpu.py:133-139
# lines [133, 135, 136, 137, 138, 139]
# branches ['135->136', '135->139']
import pytest
from unittest.mock import patch, MagicMock
# Assuming ScaleneNVIDIAGPU is in the scalene_nvidia_gpu.py file within the scalene package
# and that the pynvml module is not available, we mock the entire scalene_nvidia_gpu module.
# We also assume that the __pid attribute is set somewhere within ScaleneNVIDIAGPU.
# Mocking the entire scalene_nvidia_gpu module
with patch.dict('sys.modules', {'pynvml': MagicMock()}):
from scalene import scalene_nvidia_gpu
@pytest.fixture(scope="function")
def mock_gpu_stats():
with patch.object(scalene_nvidia_gpu.ScaleneNVIDIAGPU, 'has_gpu', return_value=True), \
patch.object(scalene_nvidia_gpu.ScaleneNVIDIAGPU, 'gpu_utilization', return_value=50.0), \
patch.object(scalene_nvidia_gpu.ScaleneNVIDIAGPU, 'gpu_memory_usage', return_value=1024.0):
yield
@pytest.fixture(scope="function")
def mock_no_gpu():
with patch.object(scalene_nvidia_gpu.ScaleneNVIDIAGPU, 'has_gpu', return_value=False):
yield
def test_get_stats_with_gpu(mock_gpu_stats):
gpu = scalene_nvidia_gpu.ScaleneNVIDIAGPU()
gpu._ScaleneNVIDIAGPU__pid = 1234 # Mocking the __pid attribute
total_load, mem_used = gpu.get_stats()
assert total_load == 50.0
assert mem_used == 1024.0
def test_get_stats_without_gpu(mock_no_gpu):
gpu = scalene_nvidia_gpu.ScaleneNVIDIAGPU()
gpu._ScaleneNVIDIAGPU__pid = 1234 # Mocking the __pid attribute
total_load, mem_used = gpu.get_stats()
assert total_load == 0.0
assert mem_used == 0.0
|
