Skip to content

[CUDA graph] Drain the device after the last warmup, before capture #8

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
jiaqiang-dot-liu wants to merge 1 commit into
base: main
Choose a base branch
from
Open

[CUDA graph] Drain the device after the last warmup, before capture #8

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
6 changes: 6 additions & 0 deletions python/sglang/srt/model_executor/runner_backend/breakable_cuda_graph_backend.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -124,6 +124,12 @@ def capture_one(
if post_warmup_hook is not None:
post_warmup_hook()

# The syncs above order work issued before each warmup, not the last
# warmup's own. A JIT module that finishes loading mid-capture issues
# illegal driver calls on the capturing stream, so drain once more here.
self._device_module.synchronize()
self._tp_group.barrier()

graph = BreakableCUDAGraph(self.deduped_cuda_graph)
captured_fn = (
eager_on_graph(True)(forward_fn) if self._debug_eager else forward_fn
Expand Down
6 changes: 6 additions & 0 deletions python/sglang/srt/model_executor/runner_backend/full_cuda_graph_backend.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -157,6 +157,12 @@ def capture_one(
self._reuse_output_buffer = self._output_buffer is not None
del warmup_output

# The syncs above order work issued before each warmup, not the last
# warmup's own. A JIT module that finishes loading mid-capture issues
# illegal driver calls on the capturing stream, so drain once more here.
self._device_module.synchronize()
self._tp_group.barrier()

graph = torch.cuda.CUDAGraph()

graph_ctx: Callable[..., AbstractContextManager]
Expand Down
Loading