[CI][sleep mode] fix sleep pytest CI error (#4912)
Signed-off-by: vensen <vensenmu@gmail.com>
This commit is contained in:
1 parent
14fad9835d
commit
ba7e9e6496
1 file changed
+8
-8
@@ -179,7 +179,7 @@ class TestOmniSleepMode:
|
||||
try:
|
||||
device_id = 0
|
||||
used_before = get_device_global_memory_used_gib(device_id)
|
||||
acks = await llm_engine.sleep(stage_ids=[0], level=2)
|
||||
acks = await llm_engine.sleep(stage_ids=[0], level=1)
|
||||
await asyncio.sleep(1.5)
|
||||
used_after = get_device_global_memory_used_gib(device_id)
|
||||
drop_gib = used_before - used_after
|
||||
@@ -209,7 +209,7 @@ class TestOmniSleepMode:
|
||||
"""Diffusion Worker stage signal loop"""
|
||||
try:
|
||||
logger.info("Starting Diffusion Worker Handshake Test")
|
||||
acks = await diffusion_engine.sleep(stage_ids=[0], level=2)
|
||||
acks = await diffusion_engine.sleep(stage_ids=[0], level=1)
|
||||
|
||||
def _get_status(ack):
|
||||
return ack.status if hasattr(ack, "status") else ack.get("status")
|
||||
@@ -264,9 +264,9 @@ class TestOmniSleepMode:
|
||||
base_output = output
|
||||
assert base_output is not None and len(base_output.images) > 0
|
||||
logger.info("Baseline Generation successful.")
|
||||
# Sleep Level 2
|
||||
# Sleep Level 1
|
||||
logger.info("Entering Deep Sleep (VRAM Scavenging)...")
|
||||
await diffusion_engine.sleep(stage_ids=[0], level=2)
|
||||
await diffusion_engine.sleep(stage_ids=[0], level=1)
|
||||
# Wake-up
|
||||
logger.info("Waking up (Reloading Weights)...")
|
||||
await diffusion_engine.wake_up(stage_ids=[0])
|
||||
@@ -353,8 +353,8 @@ class TestOmniSleepMode:
|
||||
logger.info(f"Diffusion Initial VRAM: {vram_initial:.2f} GiB")
|
||||
|
||||
# Sleep
|
||||
logger.info("Triggering Level 2 Deep Sleep (Full Weight Offloading)...")
|
||||
acks = await diffusion_engine.sleep(stage_ids=[0], level=2)
|
||||
logger.info("Triggering Level 1 Deep Sleep (Partial Weight Offloading)...")
|
||||
acks = await diffusion_engine.sleep(stage_ids=[0], level=1)
|
||||
|
||||
reported_freed_bytes = sum(getattr(ack, "freed_bytes", 0) for ack in acks)
|
||||
reported_freed_gib = reported_freed_bytes / 1024**3
|
||||
@@ -485,7 +485,7 @@ async def test_diffusion_model_sleep_tp(tp_size: int):
|
||||
async for _ in engine.generate("test", sampling_params_list=[llm_sp, diff_sp]):
|
||||
pass
|
||||
|
||||
acks = await engine.sleep(level=2)
|
||||
acks = await engine.sleep(level=1)
|
||||
statuses = [get_ack_info(ack, "status") for ack in acks]
|
||||
assert all(s == "SUCCESS" for s in statuses), f"Sleep failed. Statuses: {statuses}"
|
||||
|
||||
@@ -541,7 +541,7 @@ async def test_multistage_sleep_h100(tp_size: int):
|
||||
async for _ in engine.generate("warmup", sampling_params_list=[SamplingParams(), sp]):
|
||||
pass
|
||||
|
||||
acks = await engine.sleep(stage_ids=[0, 1], level=2)
|
||||
acks = await engine.sleep(stage_ids=[0, 1], level=1)
|
||||
assert len(acks) == 2
|
||||
|
||||
await engine.wake_up(stage_ids=[0, 1])
|
||||
|
||||
Reference in new issue
Block a user