Skip to content
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.

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

[CI] Enable 4- and 8-device Flash Attention layers tests. #1027

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open
Changes from all commits
Commits
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
9 changes: 9 additions & 0 deletions run_tests.sh
Original file line number Diff line number Diff line change
Expand Up @@ -69,3 +69,12 @@ while [ ${#TEST_PIDS[@]} -ne 0 ]; do
wait -n -p PID ${!TEST_PIDS[@]} || exit_if_error $? "Test failed."
unset TEST_PIDS[$PID]
done

# Simulate 4- and 8-device environment to run Flash Attention layer tests.
# Run at the end to avoid OOM.
for num_devices in 4 8; do
XLA_FLAGS="--xla_force_host_platform_device_count=${num_devices}" pytest \
--durations=100 -v -n auto \
-m "not (gs_login or tpu or high_cpu or fp64)" axlearn/common/flash_attention/layer_test.py \
--dist worksteal
done