Skip to content

[Reland] Fix SimpleFSDP aot_eager_autobucket #1965

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

Merged
fegin merged 2 commits into from
Oct 29, 2025
+4 −2
Merged

[Reland] Fix SimpleFSDP aot_eager_autobucket #1965

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b087f5c
Update (base update)
fegin Oct 29, 2025
f07c3cd
Update
fegin Oct 29, 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
6 changes: 4 additions & 2 deletions torchtitan/experiments/simple_fsdp/backend.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,12 +21,14 @@ def get_compile_backend(backend_name: str) -> Union[str, callable]:
# Perform auto optimization in aten fx-level and execute code in aot_eager backend
# The autobucketing logic is here: https://github.com/pytorch/pytorch/pull/163960
from torch._dynamo.backends.common import aot_autograd as aot_autograd_backend

from torch._inductor.config import aten_distributed_optimizations as dist_opts
from torch._inductor.fx_passes.overlap_scheduling import (
schedule_overlap_bucketing,
)

torch._inductor.config.test_configs.aten_fx_overlap_preserving_bucketing = True
torch._inductor.config.test_configs.aten_fx_overlap_insert_overlap_deps = False
dist_opts.collective_bucketing = True
dist_opts.insert_overlap_deps = False
torch._inductor.config.allow_buffer_reuse = False

def aten_autobucketing_reordering_pass(
Expand Down