def test_reward_space(env: LlvmEnv):
env.reward_space = "IrInstructionCount"
assert env.reward_space.id == "IrInstructionCount"
env.reward_space = None
assert env.reward_space is None
invalid = "invalid value"
with pytest.raises(LookupError) as ctx:
env.reward_space = invalid
assert str(ctx.value) == f"Reward space not found: {invalid}"
def test_apply_state(env: LlvmEnv):
"""Test that apply() on a clean environment produces same state."""
env.reward_space = "IrInstructionCount"
env.reset(benchmark="cbench-v1/crc32")
env.step(env.action_space.flags.index("-mem2reg"))
with gym.make("llvm-v0", reward_space="IrInstructionCount") as other:
other.apply(env.state)
assert other.state == env.state
def test_connection_dies_default_reward_negated(env: LlvmEnv):
env.reward_space = "IrInstructionCount"
env.reset(benchmark="cBench-v0/crc32")
env.reward_space.default_negates_returns = True
env.reward_space.default_value = 2.5
env.episode_reward = 10
env.service.close()
observation, reward, done, _ = env.step(0)
assert done
assert reward == -7.5 # negates reward.
def test_same_reward_after_reset(env: LlvmEnv):
"""Check that running the same action after calling reset() produces
same reward.
"""
env.reward_space = "IrInstructionCount"
env.benchmark = "cbench-v1/dijkstra"
action = env.action_space.flags.index("-instcombine")
env.reset()
_, reward_a, _, _ = env.step(action)
assert reward_a, "Sanity check that action produces a reward"
env.reset()
_, reward_b, _, _ = env.step(action)
assert reward_a == reward_b
def test_connection_dies_default_reward(env: LlvmEnv):
env.reward_space = "IrInstructionCount"
env.reset(benchmark="cbench-v1/crc32")
env.reward_space.default_negates_returns = False
env.reward_space.default_value = 2.5
env.episode_reward = 10
# Kill the service. Note killing the service for a ManagedConnection will
# result in a ServiceError because we have not ended the session we started
# with env.reset() above. For UnmanagedConnection, this error will not be
# raised.
try:
env.service.close()
except ServiceError as e:
assert "Service exited with returncode " in str(e)
_, reward, done, _ = env.step(0)
assert done
assert reward == 2.5
def test_set_reward_space_from_spec(env: LlvmEnv):
env.reward_space = env.reward.spaces["IrInstructionCount"]
reward = env.reward_space
env.reward_space = "IrInstructionCount"
assert env.reward_space == reward
