Skip to content

Fix: Prevent trace file overwrite across ranks in torch profiler when using multiple GPUs #9022

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 4 commits into
base: main
Choose a base branch
from
Open

Fix: Prevent trace file overwrite across ranks in torch profiler when using multiple GPUs #9022

Changes from 3 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
f0a032a
Update py_executor.py
clintg6 Nov 7, 2025
0e61d51
Update py_executor.py
clintg6 Nov 8, 2025
550c480
Update py_executor.py
clintg6 Nov 8, 2025
c67c6d6
Merge branch 'main' into fix/multi-gpu-torch-profiling
clintg6 Nov 11, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 15 additions & 4 deletions tensorrt_llm/_torch/pyexecutor/py_executor.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -504,6 +504,15 @@ def should_stop_processing(self):
return self.is_shutdown and len(self.active_requests) == 0 and \
self.executor_request_queue.get_waiting_queue_size() == 0

def _get_ranked_trace_path(self):
"""Return a per-rank torch trace path based on TLLM_TORCH_PROFILE_TRACE."""
rank = getattr(self.dist, "rank", 0)
trace_path = self.torch_trace_path
if os.path.isdir(trace_path):
return os.path.join(trace_path, f"trace_{rank}.json")
base, ext = os.path.splitext(trace_path)
return f"{base}_{rank}{ext or '.json'}"

@contextmanager
def _profiler(self):
it = -1
Expand All @@ -518,12 +527,12 @@ def _profiler(self):
start_event_2 = None
end_event_2 = torch.cuda.Event(enable_timing=True)
prev_device_step_time = None

torch_trace_path = os.environ.get(PROFILE_TRACE_ENV_VAR_NAME, None)
self.torch_trace_path = os.environ.get(PROFILE_TRACE_ENV_VAR_NAME, None)
profile_start_stop = os.environ.get(PROFILE_START_STOP_ENV_VAR_NAME,
None)
enable_torch_trace = bool(torch_trace_path and profile_start_stop)
if torch_trace_path and profile_start_stop is None:
enable_torch_trace = self.torch_trace_path and profile_start_stop
if self.torch_trace_path is not None and profile_start_stop is None:
logger.warning(
f"{PROFILE_START_STOP_ENV_VAR_NAME} environment variable "
"needs to be set to enable the torch trace. Example to profile "
Expand All @@ -546,6 +555,7 @@ def profile_step():
assert enabled, "Inconsistent CUDA profiling state"
if enable_torch_trace:
torch_profiler.stop()
torch_trace_path = self._get_ranked_trace_path()
torch_profiler.export_chrome_trace(torch_trace_path)
logger.info(f"Profiling stopped at iteration {it}, "
f"trace saved to {torch_trace_path}")
Expand Down Expand Up @@ -612,6 +622,7 @@ def profile_step():
# Stop on early exit / exception
if enable_torch_trace:
torch_profiler.stop()
torch_trace_path = self._get_ranked_trace_path()
torch_profiler.export_chrome_trace(torch_trace_path)
logger.info(f"Profiling stopped at iteration {it}, "
f"trace saved to {torch_trace_path}")
Expand Down