Skip to content

Fix x86 CPU CI OOM by capping Inductor compile threads #5930

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
gchalump wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Fix x86 CPU CI OOM by capping Inductor compile threads #5930

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
8 changes: 8 additions & 0 deletions .github/workflows/fbgemm_gpu_ci_cpu.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -203,6 +203,14 @@ jobs:

- name: Test with PyTest
timeout-minutes: ${{ matrix.host-machine.timeout }}
env:
# Serialize Inductor's C++ compile workers. By default Inductor spawns
# one worker per vCPU (16 on these runners), and the torch.compile-heavy
# TBE tests blow past the 32 GB RAM on x86 linux.4xlarge (c5.4xlarge),
# OOM-killing the runner agent ("lost communication"). The arm
# linux.arm64.m7g.4xlarge (64 GB) survives. Capping to 1 keeps peak
# memory in budget without changing test coverage or runner cost.
TORCHINDUCTOR_COMPILE_THREADS: 1
run: . $PRELUDE; test_all_fbgemm_gpu_modules $BUILD_ENV

- name: Push Wheel to PyPI
Expand Down
Loading